Listen to this episode on Apple Podcasts, YouTube, Spotify, and Substack. Transcripts are also available by clicking on the “transcripts” button in the top right corner of this post on the app/web version. You can also access them from the email version by clicking here.

In today’s episode, I’m speaking with Matt Cragun, Director of Product for Autonomous Vehicles at NVIDIA. Matt breaks down NVIDIA’s strategy for autonomous vehicle development and how the company is approaching both the hardware and software sides of the AV stack. We discuss how NVIDIA works with AV developers to accelerate development, the role of its Hyperion platform, and how NVIDIA is making its technology easier to integrate across different vehicle platforms.

We also dive deep into Alpamayo, NVIDIA’s open-source family of AI models for autonomous driving. Matt explains how Alpamayo works, how it was trained, and how developers can adapt the models to their own use cases. We discuss the limitations of the technology, how it approaches edge cases, and how Alpamayo combines end-to-end AI with more classical autonomous driving approaches rather than relying entirely on one or the other.

Matt also discusses how Level 4 autonomy will eventually make its way from robotaxis into consumer vehicles, where the AV industry stands today, and what still needs to happen before AV technology can scale more broadly. Finally, he shares what excites him most about the next few years of autonomous driving and where he expects the technology to make the biggest advances.

Chapters

• (00:00) Introduction to Matt Cragun

• (02:10) What is Alpamayo?

• (03:50) Is NVIDIA more focused on building the hardware (chips) or software (models) for AVs?

• (05:29) NVIDIA’s strategy for AV development

• (07:03) How NVIDIA helps new AV developers accelerate development

• (08:44) How open-source is Alpamayo, and can it be run on non-NVIDIA chips?

• (09:28) Hyperion and how NVIDIA democratizes access to AV hardware

• (10:40) How Hyperion is vehicle-agnostic and easy to integrate into different vehicle platforms

• (11:20) How Level 4 autonomy will make its way from robotaxis to consumer vehicles in the coming years

• (15:10) Digging deep into Alpamayo: How it works and how it was trained

• (17:16) The limitations of Alpamayo and how AV developers can tailor it to their use cases

• (19:20) The different use cases for Alpamayo, and its approach

• (22:15) How Alpamayo combines the end-to-end AI and classical AV approaches

• (27:23) How their AV models handle edge cases

• (30:46) The current state of AV development

• (32:52) What excites Matt most about the next few years for AVs

• (33:50) Conclusions and final thoughts

Notes/Links:

• You can find Matt Cragun on LinkedIn.

• An article about Alpamayo 2 Super on Hugging face, mentioned at the 04:43 timestamp (link).

• NVIDIA’s article about Alpamayo 2 Super (link).

• NVIDIA Alpamayo landing page (link).

-Harry

Top Story of the Week

Aurora Launches Second-Generation Driverless Trucks in U.S. to Meet Customer Demand (link).

The new fleet will be deployed across Aurora’s commercial network of 10 driverless routes in the U.S. Sun Belt, allowing the company to serve more customers as it expands its autonomous freight operations. The trucks replace Aurora’s first commercial fleet, which entered service last year when the company became the first to operate Class 8 trucks without a driver on public roads in the U.S. Those vehicles have since completed nearly 440,000 driverless miles. Aurora expects the new fleet to grow to 200 driverless trucks by the end of the year.

“Last year’s driverless launch proved our technology could operate safely on public roads – our new platform now provides the foundation to deliver at scale,” said Chris Urmson, CEO and co-founder of Aurora.

Other Stuff

Zoox issues software recall after a robotaxi got confused by heavy smoke (link).

The software update “enhances the existing capability of detecting active [emergency] scenes by adding the ability to detect and respond to heavy smoke in certain situations.”

Report: Waymo Cars Mean Way Mo’ Traffic (link).

Waymo cabs spend far more time driving around without passengers than comparable human-driven Uber and Lyft cabs. In California, where Waymo’s autonomous taxis have been operating and churning out trip data for years, a Waymo will travel 0.71 miles empty for every one mile. But in New York City, so-called high-volume taxis — Uber, Lyft and the like — drive empty for 0.49 miles for every one mile with a passenger. That 14.8-percent gap would translate to tens of thousands of additional vehicle miles every single day.

Related: Are Waymo robotaxis actually safer than NYC Uber and taxi drivers? (link). A lot of the safety analysis cited in this article comes from the report referenced above, but the researchers have since pulled that entire section after discovering discrepancies in the NYC crash data and definitions they used. So for now, I’d take those safety findings, including the claim that Waymo had 1.5x the serious injury/fatal crash rate of NYC for-hire drivers, with a pretty big grain of salt. The report’s authors say they’re working to resolve the discrepancies and update the analysis.

Meanwhile, there has also been some recent research from the Insurance Institute for Highway Safety (IIHS) reinforcing Waymo’s safety claim: Waymo’s driverless cars crash less often than people (link). They found Waymo’s driverless vehicles had 68% fewer police-reportable crashes and 81% fewer injury crashes per mile than human drivers across roughly 50 million driverless miles. This is consistent with Waymo’s own safety research, which found a 79% reduction in injury crashes over 56.7 million miles (link).

Trial Lawyers Lobby Against Autonomous Vehicles (link).

Remarkably the trial lawyers saw the writing on the wall very early and they have been lobbying against AVs for nearly a decade! The trial lawyers earn a huge amount litigating ordinary auto accidents–Annual U.S. auto insurance payouts (liability + PIP/MedPay) are on the order of $180–220B and trial lawyers are very eager to retain the right to sue car manufacturers for product liability.

Tesla spins up robotaxi pilots in Orlando and Tampa ahead of Q2 earnings (link).

Waymo expects D.C. robotaxi talks to continue this summer (link).

Terawatt: How EV Fleet Operators Can Turn Energy Costs Into an Advantage (link).

Hertz’s next act: From rental cars to robotaxi depots (link).

Faster Data Off the Vehicle—Faster AI on the Road

This July 27-30, Virewirx is bringing VX60 High Performance Wireless to ATS 2026. Visit Booth 6 and discover advanced mmWave networking that delivers the massive, real-time data movement that autonomous fleets and robotaxi fleets demand. Reach out to Brian Lee at info@virewirx.com for more info.

Waymo And Uber Support Bad, Incumbent-Protecting Laws In DC And NJ (link, no paywall).

Stellantis taps Mobileye for hands-free driving assist (link, no paywall).

Mobileye CEO Amnon Shashua to step aside as company pushes into robotaxis, robotics (link).

Pony.ai ties up with JD.com for China robotaxi maintenance (link).

Waymo meltdowns, Lurie pressure, spur push for tougher robotaxi rules (link, no paywall).

Under Cortese’s bill, AV companies would be required to immediately dispatch a technician to move a car if it has a steering wheel or allow first responders to move it. Each failure to move cars or ensure that first responders can move them would result in a $5,000 fine, and if an AV causes a roadblock for more than 30 minutes and blocks first responders, the fine would be $10,000. Failure to comply with the law would result in a civil action against the AV company.

Just a few days until SAE International’s Automated Transportation Symposium in San Diego (link).

‘Uber robot taxi, new competitor incoming?’ (link). Closeup of an Uber-Lucid-Nuro robotaxi.

They can do that because they operate using a Limo license like Uber Black, and not an AV permit like Waymo (link).

Related: How the CPUC Handles AV Permits, Data, and Enforcement - Pat Tsen (link).

‘Updated Ojai Cup Holders’ (link).

What else we're reading/listening to

Waymo’s Lead Is Real. Its Biggest Test Is What Comes Next (link). Our latest guest piece where former founding member of Uber’s Autonomous Mobility & Delivery team, Eduardo Rojas, analyzes Waymo’s lead in the Robotaxi space. What strikes me here is the gap between impressive operational scale and actual market penetration. Waymo is currently completing 500,000 rides per week, which annualizes to about 26 million rides. That sounds significant, but against an estimated four billion annual U.S. ride-hail trips, it represents only 0.65% of total demand.

Rethinking AV Safety as Robotaxis Scale with Junko Yoshida (link). My latest podcast with the veteran tech journalist discusses what journalists still get wrong in their coverage of AVs today, the state of the industry, what needs to happen to scale robotaxis, and why transparency remains one of the industry’s biggest challenges.

Related: I’ll be on stage with at SAE International’s Automated Transportation Symposium on July 30th at 11:40 am on the ‘Journalist Panel’, alongside ATS co-founder Jane Lappin, Alan Ohnsman of Forbes, and David Welch of Bloomberg. Feel free to reach out if you’ll be in attendance!

AVs/Humans behaving badly

‘They’re going to kill me’: A Waymo rider was trapped inside as vandals smashed the robotaxi (link, no paywall).

The car’s remote assistant had noticed the car was not moving, and decided to call the police. Watson grew frantic. He asked Waymo’s support team when law enforcement would arrive, but they didn’t know. He asked if they could move the car, and they declined, saying it was obstructed.

‘Absolutely Insane Route’ (link). I’ve had several weird routing issues with Waymo lately and it seems like I’m not the only one. Hard to tell if routing is getting worse for some reason or these are just anomalies.

‘What’s the longest you’ve had to wait for a Waymo?’ (link).

‘My Ojai re-routed into the Waymo backrooms and rider support had to step in’ (link).

Shout-outs

Big thanks to TDD readers at Element Fleet Management, Uber, and to Andy L and Mark M for referring new subscribers. If there’s someone you think would enjoy TDD, feel free to forward this email or use the referral button below, and we’ll make sure to shout out your company.

Refer a friend

Neat Jobs

Staff Product Data Scientist, Paid Media at Waymo (link) via Hongying Z.

Physical Infrastructure Engineer at Waabi (link) via Gregory Carvlin.

Senior Software Engineer, Cloud, Simulation at Wayve (link) via Arvind Pereira.

UK Operations Lead at Waymo (link) via Leticia Cavalcante.

Job Moves

Mason Lee: Nvidia -> Rivian (link).

Laura Salazar Lawton: Torc Robotics -> Avride (link).

Sendil Palani: Tesla -> Zipline (link).

Girish Venkataramani: Wayve -> OpenAI (link).

Neha Sinha: Thermo Fisher Scientific -> Waymo (link).

We’re now at 11,000+ subscribers! Interested in advertising with TDD? Check out our media kit here and reach out to us. Newsletter spots are currently sold out until September but we do have podcast advertising opportunities available.

Reach out

Until next week :)

-Harry

Waymo is the clear leader in the U.S. robotaxi race.

It is the only company operating commercial no-vehicle-operator (NVO) services across 10+ U.S. markets. Today, Waymo is completing more than 500,000 paid rides per week, and its U.S. fleet appears to be approaching 4,000 vehicles.

That lead is real, and it is not small. Its closest U.S. competitor, Zoox, reportedly has a fleet in the low hundreds of vehicles, less than 10% of Waymo’s fleet, and while Zoox is offering public rides in Las Vegas and San Francisco, the company still cannot charge customers as it awaits final regulatory approval for its purpose-built robotaxi.

Waymo built that advantage through patient capital, engineering depth and sustained execution. Alphabet’s profitability gave it room to absorb years of losses, keep investing through setbacks and pursue a problem that most companies could not afford to chase for that long. The engineering made driverless driving work; the execution turned it into a commercial product across 10+ U.S. markets and supported plans for an aggressive footprint expansion this year.

But the more interesting question is what happens next. Has Waymo built a lead that is structurally hard to catch, or has it simply won phase one of a much longer race?

I think the next 12 months will be critical to answering that question. Waymo is clearly ahead, but it has also been operating in the U.S. with limited direct competition. That is starting to change. Zoox is preparing to scale, Tesla remains the wildcard, Uber is assembling a portfolio of AV partners, and Chinese players have already shown that large-scale NVO deployments are possible.

Waymo won the first stage, but the market is still early. The next test is how it performs as the U.S. robotaxi landscape becomes more competitive.

Robotaxis vs. Ride-Hail

Waymo leads the autonomous mobility category in the U.S., but robotaxis still represent a very small part of the broader ride-hail market.

Waymo is completing roughly 500,000 rides per week, which annualizes to about 26 million rides. That sounds large because, for autonomous vehicles, it is. But against an estimated four billion annual U.S. ride-hail trips1, it represents roughly 0.65% of the market.

Even if Waymo reaches its stated goal of one million weekly rides, that would annualize to roughly 52 million rides, or about 1.3% of estimated U.S. ride-hail demand. That would be a significant milestone for the AV industry, but still only a small fraction of the broader rideshare market.

That said, Waymo already has a meaningful presence in key cities. San Francisco is the clearest example, where it has reportedly reached roughly 25% market share and is continuing to expand across the Bay Area.

But San Francisco is also Waymo’s cleanest product-market fit. It is dense, expensive, tech-forward, and filled with high-income riders who are more willing to try new technology and pay premium prices. Obi has reported that Waymo rides can cost 13–27% more than Uber and Lyft, which makes San Francisco a useful but imperfect test case. It is less clear that riders in less affluent or less tech-forward markets will make the same tradeoff, especially while Waymo’s coverage remains limited.

That is why the next phase is so important. The question is no longer whether Waymo can make driverless rides work. It can. The question is whether the company can become the default AV option in the most important U.S. markets before competitors establish a real presence. Waymo has proven that customers value the product in its strongest markets; it has not yet demonstrated that it can translate that lead into meaningful share of the broader U.S. rideshare market.

Waymo’s Lead Is Real. So Why Isn’t It Bigger?

If Waymo’s service works and consumer demand exists, why hasn’t the operator expanded faster?

I think the answer is supply.

Waymo’s fleet is still small relative to the size of the opportunity. A few thousand vehicles is a lot for robotaxis, but it is not enough to serve the largest U.S. ride-hail markets at real density. Expanding into more cities matters, but so does having enough vehicles in each market to offer short wait times, broad coverage and reliable availability.

I think its challenge has been finding the right vehicle partners. The Jaguar I-PACE helped Waymo prove the category but it was never a long-term scaling platform. Jaguar was winding down the platform, and Waymo needed a partner capable of supporting its expansion plans.

That is harder than it sounds. Traditional automakers are built around programs measured in hundreds of thousands, or millions, of vehicles. Robotaxi companies are still talking about fleets in the thousands, and eventually tens of thousands. Those volumes are meaningful for autonomous mobility but relatively small for a global automaker.

There is also a strategic tension. If autonomous mobility works, it may shift some demand away from car ownership and toward mobility-as-a-service. Automakers want to sell vehicles, while robotaxi companies want to turn those vehicles into service infrastructure. Their interests overlap but they are not perfectly aligned.

Waymo is now trying to address the supply constraint with two new platforms: the Zeekr-built Ojai and the Hyundai IONIQ 5. In a June ‘26 interview with Ben Thompson, Michael Morton from Moffett Nathanson estimated Waymo is importing Ojai vehicles at a rate of roughly 300 per month. If that pace holds, it could add more than 3,000 vehicles over the next year, roughly doubling the current fleet.

But imported vehicles are not the same as deployed robotaxis. The real test is how quickly Waymo can validate, permit, and put them into service—and whether the new platforms preserve the product experience that helped build demand.

The Ojai introduces two additional risks. Because it is built in China by Zeekr, it is subject to the 100% China-specific Section 301 tariff in the U.S., making the vehicle materially more expensive. At least on first impression, it appears less premium than the I-PACE. That may not matter to all customers, but it could become more important if Waymo wants to sustain a 10–20% pricing premium as it expands beyond San Francisco.

The IONIQ 5 creates a different challenge. Avride, Motional, and potentially other AV companies are also integrating on the same vehicle, which means the physical product itself may become less distinctive. If several operators use the same platform, Waymo will need to differentiate more clearly through the Driver, the app, service reliability, and the overall rider experience.

The new vehicles should help Waymo scale. The open question is whether they can do so without diluting the product advantage that got them here.

The Opportunity: Win The Largest U.S. Ride-hail Markets

These new vehicles give Waymo a real opportunity to expand across the most important U.S. ride-hail markets.

Waymo has already entered or announced many of the largest markets. Based on my demand-weighted view of U.S. ride-hail, it is currently live in metros representing roughly 27% of estimated demand and has announced or signaled markets representing another 43%. In other words, Waymo now has a presence—or a stated intention to establish one—across close to 70% of estimated U.S. ride-hail demand.

That does not mean Waymo serves 70% of the market today. Nor does it mean its service areas cover entire metros. This is a metro-level demand proxy, not a measure of actual coverage. I count Los Angeles as a live market because Waymo operates there, for example, even though its service remains concentrated within a limited portion of the broader metro area.

But that still matters. Waymo is already in, or heading to, many of the markets that drive U.S. rideshare demand. Now it needs to make the service meaningfully available in each one: more vehicles, shorter wait times, broader coverage, airport access, and more consistent reliability.

In robotaxis, availability is the product. A service that works great but comes with long wait times or is not available in your area is still not a true substitute for Uber or Lyft.

The key point is simple: Waymo has established the market footprint. The next 12 months are about building density within it.

That is both the opportunity and the risk. Waymo’s next chapter is less about proving the Driver and more about industrializing the fleet.

Competition Is Closing In

Waymo is not facing one obvious “Waymo killer.” The threat is broader than that. Several credible players are moving into deployment over the next 6–12 months.

Zoox is carrying public riders in Las Vegas and select riders in San Francisco, although rides remain free. Tesla’s Robotaxi service is available in limited areas of Austin, Dallas, and Houston. Avride has offered rides through Uber in Dallas since December 2025, initially with an onboard specialist behind the wheel. Nuro has secured both a California driverless testing permit and a permit to test passenger services with a safety driver. MOIA is testing autonomous ID. Buzz vehicles in Los Angeles ahead of a planned Uber launch in late 2026, while Wayve and Uber are preparing passenger trials in London.

None of these companies are at Waymo’s level today, and early deployment does not mean the technology is mature. NHTSA has opened a preliminary evaluation into 16 reported crashes involving Avride vehicles in Austin and Dallas. The crashes resulted in property damage and one alleged minor injury, and all occurred while an in-vehicle operator was supervising the system.

Zoox is the runner-up. It is probably the most credible U.S. challenger to Waymo’s rider experience, with a purpose-built robotaxi, Amazon backing, and a factory capable of assembling more than 10,000 vehicles per year at full scale. If Zoox secures the approvals needed to charge riders and expands beyond its limited footprint, it becomes a real second player in the U.S. robotaxi market.

Tesla is the dark horse. It is no longer competing only through announcements: its service is live in three Texas markets, although coverage remains limited. If Tesla can prove the product works reliably at scale, it would be difficult to compete with: low-cost vehicles, massive production capacity, a direct consumer app, a powerful brand, and a leader who can shape the industry narrative. But the rollout remains small and has included reported crashes, so this is still an execution story rather than a solved one.

Uber’s bench is the platform threat. It may not need to own an autonomous-driving stack if it controls the demand layer. By backing Nuro, Avride, MOIA, Wayve, WeRide, Pony.ai, and other suppliers, Uber is building alternatives to vertically integrated players such as Waymo, Zoox, and Tesla. Combining those autonomous fleets with Uber’s structural advantage—human drivers who can cover peaks, edge cases, and markets where AV supply remains thin—could allow it to keep wait times low while maintaining high AV utilization.

The Chinese players matter because they show that Waymo’s technology path is not unique. They are already operating across several markets in mainland China and are beginning to expand internationally. Pony.ai has launched paid robotaxi rides in Zagreb with local operator Verne, with Uber integration expected to follow. Lyft also plans to deploy Baidu Apollo Go vehicles in Germany and the UK.

The UK could become the first Western market where these companies compete from roughly the same starting line. Unlike San Francisco, where Waymo has had years to build brand recognition, operating experience, and customer habits before a serious competitor arrives, London could see Waymo, Wayve, and Apollo Go enter within a similar window. That would give us the clearest test yet of whether Waymo’s product is truly differentiated—or whether part of their advantage has simply come from being first.

Mainland Europe could become an important proving ground. Waymo and U.S.-backed providers are exploring the region, but so are European operators and Chinese AV companies. Europe’s fragmented regulatory landscape means deployment will likely happen market by market, giving new entrants multiple paths to build a safety record and commercial credibility. More importantly for the U.S. debate, if Chinese companies can prove themselves across respected European markets, that could eventually strengthen their case with U.S. regulators.

Waymo has already won the race to prove that robotaxis can work. The next race is to turn that technology lead into a durable market lead. The next 12 months will show whether it can challenge Uber and Lyft as the dominant ride-hailing platform—or whether it becomes just another autonomous fleet supplying those platforms.

Listen to this episode on Apple Podcasts, YouTube, Spotify, and Substack. Transcripts are also available by clicking on the “transcripts” button in the top right corner of this post on the app/web version. You can also access them from the email version by clicking here.

In today’s episode, I’m speaking with Junko Yoshida, a veteran technology journalist and one of the industry’s leading voices on autonomous vehicles. We discuss where the AV industry stands today, why Waymo has emerged as the clear commercial leader, and whether autonomous driving is really a solved problem, or if the industry’s biggest challenges are still ahead. Junko shares her perspective on the importance of edge cases, what the recent Fourth of July Waymo incident in San Francisco reveals about scaling robotaxi operations, and how AV companies should be judged as commercial deployments continue to expand.

We also dive into hybrid autonomy, the role of remote operators, and whether removing the safety driver is truly the right benchmark for autonomy maturity. Junko explains why transparency remains one of the industry’s biggest challenges, how journalists can better cover autonomous vehicles, and why there is often a disconnect between industry reporting and academic research. Finally, we discuss end-to-end AI, exciting developments in London’s AV ecosystem, and preview our upcoming ‘Journalist Panel’ at ATS 2026.

Chapters

• (00:00) Introduction to Junko Yoshida

• (03:30) What journalists still get wrong in their coverage of AVs today

• (05:03) Where we are in the AV landscape today and the factors still being overlooked

• (08:30) Are AVs a solved problem? Why edge cases still matter

• (10:26) Waymo’s 4th of July incident, and what they could have done better

• (17:00) Should all AV companies be judged by the same standard as Waymo?

• (17:48) Hybrid autonomy as a path to scaling AVs

• (19:00) Is removing the safety driver proof of autonomy maturity?

• (19:30) The importance of remote operators in scaling AV operations and boosting passenger confidence

• (22:40) AV regulation and how to ensure the safest deployments

• (26:29) London’s AV landscape and its key players

• (27:14) Transparency and why it’s so difficult to cover the AV industry as a journalist

• (28:30) The gap between industry reporting and academic research in the AV space

• (29:17) End-to-end AI and the potential it could unlock if successful

• (30:35) Exciting AV developments in London

• (31:40) ATS 2026: What to look out for and our ‘Journalist Panel’

• (35:38) Conclusions and final thoughts

I’ll be on stage with at SAE International’s Automated Transportation Symposium on July 30th at 11:40 am on the ‘Journalist Panel’ alongside Jane Lappin, Alan Ohnsman of Forbes, and David Welch of Bloomberg. Feel free to reach out if you’ll be in attendance!

Notes/Links:

• You can find Junko Yoshida on LinkedIn and X.

• Junko’s blog: Junko’s Tech Probe (link).

• Junko’s podcast/newsletter with Prof. emeritus : Phil & Junko on AV Safety (link).

• Waymo’s article on their remote operators ratio, mentioned at the 20:00 timestamp (link).

• Here are some of my food recommendations for ATS 2026 in San Diego, mentioned at the 32:17 timestamp:

  • Downtown/Midtown

    • Lucha Libre

      • Luchador themed taco shop. Must get the Surf and Turf Burrito with fries (California Style) and try the cilantro crema (great salsa bar).

    • Las Cuatro Milpas

      • Longtime hole in the wall spot but recently renovated. Love the homemade tortillas and rolled tacos.

    • Phils BBQ (multiple locations)

    • Bronx Pizza

    • Blue Water (middle town)

      • Great fish tacos

    • Hodad’s

  • PB/La Jolla

    • Leilani’s cafe

    • El Pescador Fish Market & Restaurant

      • Famous Fish Sandwich

  • Other

    • Lolita’s Mexican Food (Clairemont Mesa)

    • Punjabi Tandoor (multiple locations)

-Harry

Top Stories of the Week

Volkswagen enters the robotaxi race with its first passenger service launch in Germany (link).

MOIA announced Wednesday that select, preregistered Hamburg residents can now book rides in a self-driving Volkswagen ID. Buzz van through the company’s app. Up to five vehicles will operate at initial launch, with the fleet expected to expand to as many as 10 cars…Each car will have a trained safety monitor behind the wheel who can intervene when necessary

The pilot will function more like an autonomous shuttle — or a shared robotaxi — than a private ride like those offered by Alphabet’s Waymo. Passengers traveling in the same direction may share a vehicle, while pickups and drop-offs will take place at designated virtual stops.

Waymo Wants in on DC

Waymo really wants to operate in D.C. But does D.C. want Waymo? (link, no paywall).

Here’s a full recording of the hearing if you’re bored this weekend (link).

Related: Waymo says it’ll hire hundreds in D.C. if robotaxis are approved (link). A good reminder that AVs will create new jobs, though over the long run they’re likely to displace far more driving jobs than they create.

Waymo co-CEO Tekedra Mawakana: We’ve autonomously driven more than 300,000 miles in Washington, D.C., safely navigating everything from crowded evenings in Chinatown to emergency scenes in the Third Street Tunnel (link).

Other Stuff

Uber’s Autonomous Vehicle Strategy: Slow Their Adoption (link, no paywall). Well, that’s convenient. Uber’s Waymo partnership has already ended in Phoenix, and most of its other AV partners are still years away from meaningful scale. Slowing robotaxi adoption would certainly give the company more time for its broader AV strategy to catch up.

In New Jersey, a lobbyist representing Uber took the strategy a step further, circulating legislative language that would, for a period of three years, require any platform offering driverless ride-hailing services to have human drivers serve 85 percent of its rides.

Related: Uber’s CFO responds to regulatory capture claims (link).

Bliq wins approval to test driverless cars in Finland (link).

The approval makes Finland the second EU member state after Estonia to permit Bliq’s vehicles on public roads.

Related: Europe’s First Driverless Vehicle Approval with Bliq CEO Julian Glaab (link).

6 in 10 Gig Drivers Borrow to Make Payday Amid Rising Gas Prices and AV Uncertainty, New Report Finds (link).

Everee…released its 2026 Driver Workforce Report, a nationwide survey of 413 U.S. professional and gig drivers examining how rising fuel costs, pay timing, and the expansion of autonomous vehicles are affecting driver earnings and retention… 52.5% are concerned that autonomous vehicles will reduce the number of driving jobs available to them, as services like Waymo and Tesla Robotaxi continue expanding into new U.S. markets.

Restaurant delivery robots return to D.C. with AI upgrade (link).

Coco is launching about 25 autonomous robots serving Logan Circle, and wants to expand into Dupont Circle and beyond if demand is there.

Built for Autonomous Fleet Deployment

Voltera develops, owns, and operates high-capacity charging infrastructure purpose-built for autonomous vehicle fleets. By managing site selection, power coordination, construction, and operations, Voltera enables rapid deployment in dense urban markets where land, power, and permitting are the biggest constraints. As autonomy scales, infrastructure becomes the bottleneck. Voltera delivers high-performance fleet hubs that remove upfront capital and execution risk—enabling faster market entry and scalable deployment.

Please reach out to learn more about partnering with Voltera.

Contact Voltera

Had a great time celebrating the launch of Chip yesterday! Chip is a new all-electric “life utility vehicle” built for neighborhoods, campuses, resorts, and anywhere you don’t necessarily need a full-sized car, including places you do. One of the more interesting parts is Chip Go, a remote-driving service where a real human can remotely drive the vehicle for things like parking, pickups, or errands today, with the long-term goal of Level 4 autonomy in the future (link).

Avride is officially on the ground in Miami! In partnership with Uber Eats (link).

Shared Roadways, Platforms, and Benefits: Integrating Autonomous Travel Choices - The Eno Center for Transportation (link).

‘This is a genuinely interesting research survey to share your ideas for how Waymo vehicles can collect data to enhance urban environments’ (link).

Autonomous Fleets May Be Self-Driving, But They’re Not yet Fully Self-Reliant (link).

Sharing economy in the era of full automation: Evidence from autonomous vehicle on-demand mobility services (link). This research paper explores what a future city could look like with a mix of company-owned and privately owned autonomous vehicles sharing the road. Using Chicago as a case study and ridehail data from November 2024.

Waymos are everywhere in much of LA. But how many people are taking them? (link).

Though people felt less safe in Waymos compared to cars with drivers in them, that perception flipped when people were asked about how safe they felt from harassment — at least for one major demographic. “Women feel a lot safer in a Waymo compared to an Uber or Lyft,” Thomas said.

ROCSYS has been named as one of TIME’s Top 250 GreenTech Companies (link).

Ojai back seat bike storage (link).

What else we're reading/listening to

Ride AI: Ground Truth vol.3 | Building the “Air Traffic Control” for AVs with Autolane (link).

Why Are Waymos Still Blocking Bike Lanes? (link). Our latest guest piece where takes a closer look at Waymo’s repeated incursions into bike lanes, what they mean for cyclists, and how the company’s behavior compares to human rideshare drivers, despite AVs arguably being held to a higher standard. He also explores why Waymo is incentivized to keep operating this way, and what cities could do to encourage safer curbside behavior.

Building Autonomous Vessels for Cleaner Waterways - Michael Arens, Clean Earth Rovers CEO (link). My latest podcast where we explore a very unique application of autonomy: cleaning debris from waterways, highlighting how versatile autonomy is beyond robotaxis.

AV Events

On Thursday, July 23, MobilitySF is hosting Chris Yu of ALSO and Alicia John-Baptiste for a conversation on urban mobility, smaller vehicles, better streets, and the future of cities. Rivian SF Showroom, Hayes Valley. RSVP here (link).

The Austin edition of our Urban Autonomy Summit, happening on October 14, is shaping up to be a gathering of some of the brightest minds in the AV industry, all coming together in one of the most tech-forward cities in the US. Our Detroit edition was a success and we’re bringing that same energy to Austin. Apply now to secure your spot before registration fills up (link).

‘Excited to head back to SF for [TaskUs] Forward ‘26. The conversations we had with AV and robotics leaders last year were incredible, and this year is going to be even better now that everyone is talking less about promises and more about the real operational engine required to run the technology’ (link). It’s coming up on October 28, 2026, at Above the Rim at Chase Center in San Francisco, CA (link).

‘40 30 days until #Actuate26! The full speaker lineup is here!’. Some notable speakers include Wayve CEO Alex Kendall, Aurora CEO Chris Urmson, and NVIDIA VP of AI Research Sanja Fidler. The event will take place August 18–19, 2026, at Fort Mason in San Francisco (link).

‘On July 22, Signal. returns to Kodiak HQ with Kasper Sage, Managing Partner at BMW i Ventures. Join Kasper and Kodiak CEO Don Burnette for a conversation focused on how the best bets in autonomous technology are made, what investors look for, and what separates the companies that get funded from those that don’t’ (link).

AVs/Humans behaving badly

Shirtless man vandalizes Waymo in East Hollywood (link).

‘Trusted Testers, what are y’all doing???’ (link). Waymo may want to re-think cloth seats in the Ojai.

No, a Waymo did not hit a bicyclist in Miami. It was a human driver, police say (link). While police confirmed the Waymo wasn’t involved in the collision, I’d still love to see the video because I’m not entirely clear on what actually happened.

Miami Beach Police have debunked an Instagram video posted by a popular “citizens journalism” account claiming that a Waymo hit a bicyclist on Tuesday in South Beach. Miami Beach Police spokesperson Officer Christopher Bess says the driver of a Mercury Grand Marquis was parked on the side of the road and hit the cyclist when trying to make a U-turn. The driver “was determined to be at fault and was cited accordingly,” Bess said.

Robotaxi Riders Are Falling Asleep, Sparking Frantic 911 Calls (link, no paywall). Feels like something the operators need to solve. First responders shouldn’t have to spend time dealing with sleeping robotaxi passengers.

AVs behaving “goodly” :)

‘Roads are unpredictable. People run red lights, drift out of bike lanes, step off curbs without looking. The question isn't whether it happens. It's whether the vehicle is ready when it does’ (link).

Kodiak: Three trailers. 137 tons. Nearly half a football field in length… Driverless (link).

Collision Involvement Note (link).

Shout-outs

Big thanks to TDD readers at Ono, and to Mark G for referring new subscribers. If there’s someone you think would enjoy TDD, feel free to forward this email or use the referral button below, and we’ll make sure to shout out your company.

Refer a friend

Neat Jobs

Senior Strategic Operations Associate, Autonomous Mobility & Delivery at Uber (link) via Ana Loibner.

Senior Staff / Principal AI & Compute Architect at Zoox (link) via Zoox.

Strategic Partnerships Development Manager at Waymo (link) via Matthew Fisher.

Head of Product at Nuro (link) via Jiajun Zhu.

Mobility Manager at Metropolis in Los Angeles and New York, via Robin Riedel.

System Disruption Manager at Waymo (link) via Jackie Lindsey.

Job Moves

Jeff Beck: Honda -> Waabi (link).

Ryan Browne: Wayve -> ❓ (link).

Waqas Malik: Valeo -> Nuro (link).

We’re now at 11,000+ subscribers! Interested in advertising with TDD? Check out our media kit here and reach out to us. Newsletter spots are currently sold out until September but we do have podcast advertising opportunities available.

Reach out

Until next week :)

-Harry

In 2023, Christopher White, the executive director of the San Francisco Bicycle Coalition (SFBike), received a phone call from a longtime donor who was threatening to yank his funding.

At the time, Waymo was seeking approval to operate its driverless taxis around the clock in San Francisco despite opposition from local officials. The donor was frustrated that the bicycle coalition had declined to endorse Waymo’s expansion. Self-driving cars, he argued, would ultimately make cycling safer because they caused fewer serious crashes than human drivers. After a heated conversation, he invited White to experience a Waymo ride firsthand.

For their joint Waymo trip, the donor selected a dropoff location on Cesar Chavez Avenue, a San Francisco arterial where a painted cycle lane is sandwiched between two lanes of car traffic and a row of on-street parking. “It’s one of our least favorite streets for biking,” White recalled recently. “He wanted to show me how well the robotaxi interacted with bike infrastructure.”

The dropoff did not go according to the donor’s plans. Upon reaching the requested destination, the Waymo “just pulled over into the bike lane,” White said. “I was mortified because somebody on a bike actually had to swerve out of our way into fast-moving traffic.”

Three years later, Waymos are still entering bike lanes when conducting pick ups and dropoffs (examples are here, here, and here). The issue became a point of contention in the District of Columbia during a recent hearing about robotaxis, after which Ethan Teicher, a Waymo spokesperson, posted on X that “riders in places like SF will tell you Waymo vehicles will frequently drive down the block or around the corner to find an appropriate spot.” White was skeptical about that assertion; he said that his members have observed Waymos entering bike lanes “frequently enough to cause concern.”

Beyond being illegal, such maneuvers endanger anyone traveling on two wheels. Last year a San Francisco cyclist filed a lawsuit claiming that she crashed into a Waymo that was blocking a bike lane – and that she then landed on a second Waymo that was also blocking that same bike lane.

This is not a good look for a company that stakes its reputation on the idea that its vehicles save lives. But it’s also true that human drivers pause in bike lanes all the time. If Waymo refused to do so when collecting or discharging a passenger, it could cede market share to competing services like Uber and Lyft that have no such restrictions.

Waymo’s ongoing bike lane controversy—which recently went viral on social media and was covered in outlets including San Francisco Chronicle and Futurism—highlights a tension in robotaxi operations: When autonomous vehicle companies face a choice between road safety and passenger convenience, what will they do?

At the risk of stating the obvious, the purpose of a bike lane is to separate people on bikes from big metal boxes that are moving much faster. A key tool in cities’ efforts to reduce bicyclist crash deaths (which hit an all-time record in 2024), bike lanes also confer a sense of security that can entice more people to start riding.

But those benefits dissipate if motor vehicles impede cyclists’ progress. Harried drivers can view bike lanes as tantalizingly open street space, particularly in places like San Francisco where parking is in short supply. Dedicated pick up/dropoff zones can help, but cities have installed relatively few of them since ridehail emerged almost two decades ago.

A driver seeing a curb full of parked cars may conclude that the bike lane is their next-best option. After all, they will only be there for a moment, and a cyclist can always go around.

Such thinking is deeply problematic. For one thing, it implies a double standard that was satirized in 2022 by a group of activists who gathered on San Francisco’s Valencia Street, a thoroughfare notorious for bike-car conflicts. When a driver halted in a bike lane, the activists dashed into the street and waved signs with apologetic messages—“Hang tight, we’ll be just a minute!”— that cyclists often hear.

The underlying issue is a serious one, said Ken McLeod, policy director at the League of American Bicyclists, a national association for people who cycle. “If a bicyclist comes upon a car in a bike lane, they may have to go into traffic that is going by at a very high rate of speed and a heavy volume.” Doing so is uncomfortable and potentially unsafe. If a passenger is inside the vehicle, there is also an added risk of them “dooring” a cyclist as they exit (which allegedly happened to the aforementioned San Francisco cyclist who is currently suing Waymo).

Because bike lane incursions are so dangerous, many (though not all) states have banned them. For instance, California Vehicle Code Section 21209 states that “no person shall drive a motor vehicle in a bicycle lane” unless they are about to turn, preparing to exit or enter a roadway, or intend “to park where parking is permitted.” Even so, ticketing is rare. Police are unlikely to witness—let alone penalize—an infraction that lasts only a minute or two. A driver may conclude that the time they save from blocking a bike lane is worth more than the risk-adjusted punishment. Such thinking may be rational, but it invites driving behavior that is hazardous and illegal.

So what does all of this mean for a robotaxi company? Waymo did not respond to a request for comment, but last December Chris Ludwick, a senior director of product management, offered insight into Waymo’s perspective when he told the Wall Street Journal that the company’s vehicles are programmed to violate the law in certain situations that require it, such as by crossing a double-yellow line to bypass a delivery van in the street. Fair enough; any reasonable driver would do the same. The only alternative is to remain stuck indefinitely.

But that’s a very different situation from a Waymo passenger requesting a pickup or drop off at a location with a bike lane out front. “Ninety-nine percent of the time there is somewhere just a little less convenient for them to pull over that would be far safer,” said White, of SFBike. Waymo doesn’t need to block bike lanes; doing so is a choice. Although Teicher, the company’s spokesperson, suggested that its vehicles avoid such conflicts whenever possible, White said that he has seen no data supporting that claim.

Of course, the company has a powerful incentive to treat bike lanes as auxiliary parking. If Waymo did not allow its riders to be picked up or deposited where they prefer, it would confer a modest but real advantage to the Uber, Lyft, and taxi services whose drivers willingly disregard the painted lanes on the street. Especially given the immense pressure Waymo faces to scale revenue, alienating customers is not an appealing prospect.

Still, based on his own observations White believes that Waymos may be a bit less likely to stop in bike lanes than Uber and Lyft drivers, who are notorious for doing so. If that is true (and anecdotal data seems to be the only kind available), Waymo defenders might reasonably argue that the company deserves credit for improving upon human driving behavior.

But a defense that “we don’t enter bike lanes as often as Uber and Lyft” seems to place the bar rather low, particularly for a company insisting that it sets a new and higher safety standard. On its website as well as network TV, Waymo claims that its vehicles reduce serious injuries by 94 percent (an assertion that some transportation and robotics experts contest).

These kinds of tradeoffs between street safety and passenger convenience are not unique to bike lanes. Waymos have also obstructed crosswalks, which is hazardous to pedestrians and a barrier to the mobility of those in wheelchairs. McLeod expressed unease about the juxtaposition of self-driving companies’ safety claims with their seemingly deliberate violations of certain rules. “We could have automated vehicle companies saying, ‘We are safer than humans, regardless of our compliance with traffic laws. Therefore, these traffic laws are unnecessary.’ I think that is a dangerously slippery slope.”

For now, Waymo is cultivating ties with street safety communities through its sponsorships of groups like the Vision Zero Network and the Governors Highway Safety Association. That outreach includes cycling organizations as well. Last year the company flew McLeod and other staff from the League of American Bicyclists to Los Angeles to participate in demos and the recording of a slick video. (White said that Waymo previously sponsored SFBike, but that the company ceased doing so in 2023 after the organization did not endorse its push for citywide robotaxi service. He added that his members are roughly evenly split on Waymo: Some think the companies’ vehicles make cycling less dangerous, while others worry they degrade all forms of mobility that do not involve a car.)

The good news is that enhancing the safety of cyclists and other vulnerable road users need not require Waymo to go against its own interests. “Hardening” bike lanes with fixed barriers can prevent cars from entering them in the first place (research suggests they also attract many more cyclists). Santa Monica recently followed the lead of cities including Pittsburgh and Chicago by deploying automatic cameras to issue citations to cars that block bike lanes, negating the need for police enforcement.

Even Waymo has voiced support for “efforts to further protect bike lanes” from incursions by motor vehicles. Rather than cajole drivers—whether human or robotic—to stay out of bike lanes, a better strategy might be forcing them to.

-David

Listen to this episode on Apple Podcasts, YouTube, Spotify, and Substack. Transcripts are also available by clicking on the “transcripts” button in the top right corner of this post on the app/web version. You can also access them from the email version by clicking here.

In today’s episode, I’m speaking with Michael Arens, Founder and CEO of Clean Earth Rovers. We discuss how Michael went from an early failed prototype to building an autonomous maritime robotics company focused on cleaning waterways and collecting environmental data. Michael explains how Clean Earth Rovers’ autonomous vessels operate, the types of debris they collect, and the different deployment models they use for ports, marinas, municipalities, and other customers.

We also dive into the company’s approach to autonomy, including where human operators remain in the loop, why a hybrid human-autonomous model makes sense for maritime applications, and how the hardware stack compares to what’s used in robotaxis. Michael shares his perspective on the broader maritime autonomy landscape, why the economics differ from autonomous driving, and where he sees the biggest opportunities for automation on the water.

We also discuss the growing role of environmental data in Clean Earth Rovers’ business, how remote operations could evolve over time, and what the future holds for autonomous systems beyond roads and into waterways.

Chapters

• (00:00) Introduction to Michael Arens

• (01:40) What inspired Michael to start Clean Earth Rovers

• (04:10) From a failed prototype to Clean Earth Rovers

• (04:50) Their Rover’s specifications, how it works, and its use cases

• (06:44) Where Clean Earth Rovers operates, and how debris migrates there

• (09:00) How the Rovers are deployed, and the different operating models for various clients

• (11:10) How autonomous are the Rovers, and where human intervention comes in

• (12:20) What kind of debris do the Rovers collect, and how is it handled?

• (17:00) Advantages of the Human-Autonomous hybrid approach

• (21:15) The hardware stack of the Rovers, and how they compare to robotaxis

• (25:30) The maritime autonomy landscape and the innovations popping out of it

• (28:20) The value of autonomy in the maritime industry

• (30:34) The data side of the business and how they use the collected data

• (35:22) How remote operation and autonomy would evolve for Clean Earth Rovers

• (39:18) Conclusions and final thoughts

Notes/Links:

• You can find Michael Arens on LinkedIn.

• Clean Earth Rovers website (link).

-Harry

Top Stories of the Week

Waymo cars stranded on San Francisco streets as Fourth of July gridlock drains batteries (link). There are a bunch of videos from this incident, including one where a Waymo drove over a firework, and another where a gentleman counted nearly 20 stuck Waymos in a row.

It’s been a few days since this incident and Waymo still hasn’t shared more details about what happened. How many cars actually got stuck? How many had to be towed because they ran out of battery? Why weren’t they able to make it to chargers? And what are they changing so this doesn’t happen again (again)?

This also comes just a few months after the PG&E outage in San Francisco left a number of Waymos stranded. Different cause, but a similar outcome.

Here’s what Waymo had to say immediately following the incident:

“Major traffic disruptions, a high volume of travelers, and unplanned road closures contributed to unexpected congestion,” a company spokesperson said in a statement. The spokesperson said Waymo’s roadside assistance team worked to clear the vehicles, adding that the company is “evaluating ways to strengthen Waymo’s resilience in major traffic disruptions.”

Other Stuff

Tesla Robotaxi now available in Miami (link), and they say it’s unsupervised (link).

Tesla’s Robotaxi Push Tests New Blueprint for Scaling Fast (link, no paywall). Good recap, but it leaves out one important detail: Tesla’s driverless fleet in Austin is reportedly down to around 14 active vehicles, with roughly 50 Robotaxis being deployed overall. That’s still tiny compared to Waymo, which has around 250 vehicles operating in the same market.

Waymo to start driverless rides in 4 more U.S. markets as expansion accelerates (link).

Alphabet’s robotaxi division will soon start offering fully autonomous rides in San Diego, Las Vegas, Tampa, Florida, and Denver, the company said Wednesday. While the launch will initially be for Alphabet employees, the rollout will then extend to the public.

Nexar and Nauto to Merge, Creating the Independent Infrastructure Platform for Physical AI (link).

Waymo registers entities in France, the Netherlands, Spain and Germany (link).

Feds demand autonomous vehicle companies stop interfering with first responders (link).

“Let me be clear: the inability to detect and appropriately respond to such situations represents a functional insufficiency,” Morrison’s letter reads. “Emergency scenes are not rare or extreme ‘edge cases.’ As such, NHTSA is today issuing a call to action for AV developers and operators to immediately focus their resources on fixing this issue.”

The Human Edge in AV: Scaling Swift Emergency Response

TaskUs partners with the top autonomous vehicle (AV) and AI organizations to manage the human-critical workflows that make advanced systems safer and more reliable. When one of the leading companies deployed its fleets in new cities, they also needed to scale emergency response teams. TaskUs helped by deploying expert operators, creating 24/7 incident frameworks and implementing a unique fatigue management program. Get all the details and results in their case study below.

Check it out here!

One robotaxi player has grown its share of users so far in 2026 — and it’s not Waymo (link, no paywall).

‘FSD would be twice as useful in neighborhoods if I could actually talk to the car and tell it which driveway to pull into, the same way I would with a person driving me home’ (link). Tesla says they’re working on it.

‘Waymo has launched freeways in Austin on the Uber app!!’ (link). This may be a sign that freeway rides are coming back soon.

The robotaxi law that could ban Tesla (link, no paywall). Happy to see Rani Molla at a new outlet. This reminds me though of Lyft’s stance and what I wrote at the time: “I’d rather see an outcome-based framework that judges systems on real-world safety performance, not the specific technology or sensor stack they use.”

A bill expected to come up for a vote later this year would require companies seeking to operate fully autonomous vehicles in New Jersey to use cameras plus two other sensing technologies, most commonly lidar and radar… The measure would also effectively prevent Tesla’s camera-only Robotaxi system from operating in New Jersey unless the company changed its hardware.

Consumers Embrace Driverless Delivery, But Infrastructure Lags (link, no paywall).

‘I’ve lived long enough to get rug pulled by an autonomous vehicle’ (link). It’s hard to compete with Uber’s density and ETAs, but Waymo can stand out with instant matching and ETA accuracy. The typical “Waymo experience” feels especially awesome because Uber’s matching and pickup algorithm can be so inconsistent and is one of the top complaints riders have. Thus, Waymo really needs to avoid developing that same reputation with situations like this.

I used a self-driving car for 100 days. These are the mistakes it kept making (link). I was surprised by how many issues the author experienced over 100 days. It’s a much more negative experience than you typically hear from Tesla FSD supporters, though it’s hard to know how representative either perspective is. The best way to form an opinion is probably to try it yourself.

We recorded more than 500 safety-critical events where the system required driver intervention or revealed an important limitation in how it interpreted the road environment. And In more than 100 days of daily testing, we did not complete a single trip in which FSD drove independently from start to finish without intervention.

Not All Miles are Equal: Why Time and Location Matter When Benchmarking Autonomous Safety (link).

Rivian R2 spotted with LiDAR near HQ, looks better than most LiDARs (link).

Robotaxi Companies Need To Stop It With The Right Seat Safety Drivers (link, no paywall).

What else we're reading/listening to

The AV Market Strategist by : Waymo’s Grand European Expansion & The Waymo-Uber Break-Up Begins in Phoenix (link).

The Real Impact of Autonomy Will Come From Buses (link). Our latest guest piece where analyzes the impact of autonomy on buses, the opportunities it unlocks, and what the industry still needs to get right.

Einride’s Autonomous Freight Strategy with CTO Henrik Green (link). My latest podcast where we discussed how Einride is redefining autonomous freight with their cabless autonomous truck design, their hybrid network of autonomous and human-driven vehicles, and how the company combines electric trucking, autonomy, and logistics into a single platform, among other things.

AV/Humans behaving badly

Waymo calls San Mateo police on teens for drinking, shooting toy guns from driverless vehicle (link). Good one Waymo!

‘First time getting in an Ojai and seeing the front screen not working’ (link). Seems like there are still quite a few bugs with the Ojai despite them testing it for what feels like forever.

Asleep at the wheel: BC driver caught sleeping in self-driving Tesla on busy highway (link).

Shout-outs

Big thanks to TDD readers at CharterUP, and to Mike F for referring new subscribers. If there’s someone you think would enjoy TDD, feel free to forward this email or use the referral button below, and we’ll make sure to shout out your company.

Refer a friend

Neat Jobs

City Fleet Manager - Autonomous Vehicle at Freenow (link) via Swagat Chopra.

Solutions Architect at Wayve (link) via Masa Kameyama.

Program Manager, Driver and Vehicle Compliance at Waymo (link) via Karen Isgrigg.

Autonomous Vehicle Partnership Manager at Lyft (link) via William Lynn.

Principal Engineer Tech Lead at Motional (link) via Gayatri K Sukumar.

Job Moves

Balajee Kannan: Motional -> May Mobility (link).

We’re now at 11,000+ subscribers! Interested in advertising with TDD? Check out our media kit here and reach out to us. Newsletter spots are currently sold out until September but we do have podcast advertising opportunities available.

Reach out

Until next week :)

-Harry

Public transit is the backbone of cities. In New York, buses carry around 1.3 million trips a day, more than ride-hailing and taxis combined at roughly one million. In London, the gap is even wider: about five million bus journeys versus approximately 0.6 million taxi and ride-hailing trips1.

Yet most of the autonomous conversation still centers on cars. Partly because the commercial upside is obvious and near-term, and partly because cars are sexier. But it is buses that carry far more people, and that’s where autonomy will have the real impact.

More Public Transport on the Same Budget

Turning buses autonomous holds real advantages. The first is solving the driver shortage. In 2023, the Geneva-based International Road Transport Union (IRU) estimated that Europe was missing around 100,000 bus and coach drivers, equivalent to 10% of the total workforce. By 2028, that number could more than double.

Then comes the business case. Removing the driver changes the economics: lower operating costs; longer service hours, including extended nights and weekends, which raise vehicle utilisation; and smoother, more consistent driving, which can cut fuel and maintenance costs by at least 7%.

Last, the social benefits. Greater efficiency and availability could make it viable to run services that are uneconomic today, especially in lower-density areas. Add to that the potential safety gains autonomy could deliver.

The Ecosystem Is More Robust Than It Seems

Across the world today, more than 30 autonomous bus and shuttle projects are running on public roads. As of June 2026, services were operating across Europe (Norway, Sweden, Finland, France, UK), Asia (Japan, South Korea, China) and the US (Atlanta, Jacksonville), to name a few. A smaller but important subset has moved into fare-collecting services, where passengers pay and the bus is treated as part of the transport network.

The supplier landscape is taking shape too. ADASTEC, Imagry, Fusion Processing and TIER IV are among those working on autonomous bus platforms. Add shuttle players such as Ohmio, WeRide and HOLON, and the market starts to look less experimental than many assume.

Technology still needs to improve, but it is not THE limiting factor. Buses have one important advantage over robotaxis: they run on fixed routes. That creates constrained operational design domains, which can be mapped, tested and prepared for in detail. Nor do operators need to automate the whole network at once. They can go route by route, starting with the simplest corridors: bus lanes, lower speeds, easy terrain and clear road markings.

In April 2026, Norway went a step further: on a regular scheduled bus route, it removed the safety driver. Yet the driver is still on the bus, sitting in a passenger seat. That’s because drivers serve many more functions than simply drive, and removing them is where the hardest problem lives.

Respect the Driver

A driver does much more than drive. They quietly handle the many small moments that keep a service running: elderly riders who need time to board and sit down, wheelchair users who need the ramp deployed, passengers who rely on a visible human presence when things turn uncomfortable or unsafe. They are also the default answer to “what happens if…” - the accident, the breakdown, the medical emergency, or the passenger in distress. And they are part of the reason people pay. With no one at the front of the bus, how do you ensure fares are collected?

Public transport is designed to serve everyone, and vulnerable riders often lean hardest on that human presence. Remove the driver, and these tasks do not disappear. They have to be redesigned into the service. Bus autonomy is not “just” a technology challenge; it is mainly an operating model challenge.

Show Me the Money

The second major challenge is procurement. Public transport is not bought like consumer technology, where a better product can spread quickly once users like it. Buses are procured through long, risk-averse public processes shaped by central and local governments. Operators never procure at scale without a clear public budget.

That creates a vicious cycle. Few authorities are ready to budget for scaled autonomous operations, which means operators won’t procure them. That, in turn, makes it harder for technology providers to raise capital, finance development and execute deployments. And unlike Waymo, bus autonomy companies cannot simply fund and run large fleets themselves while the market catches up. So the proposition stays nascent, while everyone waits for someone else to go first.

Digital Bus Driver

As for the major challenge, the driver, the industry is already at work. The answer has two parts.

The first is service design. There will not be one driverless model. Different routes and communities will need different levels of human presence. Some services may run fully autonomously; others may carry an onboard attendant to assist passengers, at least during certain hours or on certain routes. A service might run unstaffed during the day, then add an attendant at night. Remote support can flex too: from a single control centre in another city to local “rider-assistance stations” that can reach passengers at short notice.

The second is the emergence of a new operating layer. Companies such as Beti, Applied Autonomy and Beep are rebuilding the functions that used to sit with the driver. Remote supervisors can operate the air conditioning, open and close doors, and, where permitted, intervene in how the vehicle operates. Microphones, speakers and help buttons connect passengers to the control centre, while cameras and software flag anything from anti-social behaviour to fare evasion. These tools make up the digital bus driver: a bundle of tools, software, procedures and human support channels that keep the service running.

Looking Ahead: A Slow Transition to Commercial Services

For now, autonomous bus deployments will keep relying on public funding, national innovation programmes, operator sponsorship and regulatory exemptions. But over the next five years, expect the market to shift from experimentation to early deployment. Trials will multiply, technology and digital-driver tools will mature, more services will turn commercial, and more safety drivers will move out of the driver’s seat - either into passenger-support roles or remote operations.

Watch Japan, Switzerland and Scandinavia in particular: the driver shortage is acute, regulation is moving and trials are most advanced.

Cars will keep the spotlight. But if autonomy is going to change how cities move, it will do it on the bus. Safely, reliably and affordably, on a wet Tuesday night, with real passengers, real fares and no driver at the front.

-Barak

Listen to this episode on Apple Podcasts, YouTube, Spotify, and Substack. Transcripts are also available by clicking on the “transcripts” button in the top right corner of this post on the app/web version. You can also access them from the email version by clicking here.

In today’s episode, I’m speaking with Henrik Green, Chief Technology Officer at Einride. Henrik shares his journey from a long career in the traditional automotive industry to helping lead one of the most unique companies in autonomous freight. We discuss the differences between building vehicles for consumers versus commercial autonomy, what attracted him to Einride, and why he believes the AV industry is entering an exciting new phase.

We also take a deep dive into Einride’s business model, including its cabless autonomous trucks, hybrid network of autonomous and human-driven vehicles, and how the company combines electric trucking, autonomy, and logistics into a single platform. Henrik explains the technical and economic realities of Einride’s purpose-built vehicle design, how the company approaches scaling deployments, and the regulatory considerations unique to autonomous freight.

We also explore some of the biggest technical challenges facing the industry today, Einride’s hardware choices and operational design, mapping, AI, and how they handle edge cases in real-world deployments. Finally, Henrik shares his outlook on the future of autonomous freight, where he sees the industry over the next five years, and what he’s most excited about as commercialization continues to accelerate.

Chapters

• (00:00) Introduction to Henrik Green

• (02:33) Henrik’s background in the automotive industry

• (04:00) The differences between the B2C automotive industry and the AV industry

• (05:09) What drew Henrik to Einride

• (05:50) Trends in the AV career space

• (06:50) Einride’s background, and what makes them stand out in the industry

• (07:35) Einride’s footprint, and their focus

• (10:40) Einride’s cabless autonomous trucks, their specs, and current applications

• (13:11) How Einride plans to scale, and their goals for the next year

• (13:50) The economics of Einride’s autonomous trucks, and technical specs

• (15:29) What hardware/software is unique to their cabless autonomous truck design?

• (16:30) Regulatory requirements for autonomous freight vs robotaxi operations

• (17:12) Hardware/operational differences: autonomous freight vehicles vs. robotaxis

• (19:12) How Einride’s autonomous truck form factors could be optimized in the coming years

• (20:14) Einride’s competitive advantage: Hybrid network of human-driven and autonomous freight vehicles

• (22:50) Biggest technical challenges involved in building and operating autonomous freight vehicles today

• (23:30) How Einride’s autonomous freight trucks are trained for deployment, and edge cases

• (25:48) Pre-mapping vs. end-to-end AI: how Einride uses a hybrid approach for safe deployments

• (28:40) Balancing commercialization with scaling: Einride’s approach

• (31:00) What are you most excited about over the coming years

• (34:55) 5-year outlook for the autonomous freight and robotaxi industry

• (36:24) Conclusions and final thoughts

See You at ATS 2026

The SAE Automated Transportation Symposium (ATS) is one of the premier gatherings for the autonomous vehicle industry, bringing together engineers, researchers, regulators, policymakers, and executives from across the ecosystem. I’ll be speaking on the Journalist Panel on Thursday, July 30 at 11:40 a.m. This year’s keynote speakers include Uber’s Head of Autonomous Mobility & Delivery, Sarfraz Maredia, Zoox Co-Founder and CTO Jesse Levinson, along with other leaders helping shape the future of automated transportation. Drop me a note if you’ll be in attendance, and registration is still open if you haven’t signed up yet. I hope to see you in San Diego.

Register now

Notes/Links:

• You can find Henrik Green on LinkedIn.

• Einride’s website (link).

• You can find details/pictures of their cabless autonomous trucks here.

-Harry

Top Stories of the Week

Uber and Waymo Have Ended Their Robotaxi Tie-Up in Phoenix (link, no paywall). We knew the relationship was faltering, but this is the first concrete chip to fall. Austin and Atlanta are next in my opinion. It’s also interesting that we never heard Uber mention just how small the Phoenix deployment was until after the partnership ended.

Phoenix “was an intentionally limited deployment, reaching just over a dozen vehicles dedicated to the program,” an Uber spokesperson said. Waymo said the vehicles will be integrated back into its fleet to serve a delivery agreement with Uber rival DoorDash Inc. and a public transit deal with Via Transportation Inc. that began last year. Riders in Phoenix can still hail a Waymo on the Waymo app.

Related: Is the Uber x Waymo Partnership Coming to an End? (link).

Other Stuff

Wayve launches $85M employee tender offer at $8.5B valuation (link).

Wayve.. is allowing its employees to sell a portion of their vested equity. The $85 million tender offer — essentially a structured opportunity for employees to sell shares back to investors — is being led by the company’s existing and new investors at the company’s latest valuation of $8.5 billion.

Related: The $8.6B Startup Trying to Beat Waymo Without Maps | WSJ (link).

China’s Momenta kicks off Hong Kong IPO, targets up to $751 million (link).

Momenta sells driving-assistance software to automakers. The systems help cars steer, brake, change lanes and park, but drivers still need to stay alert and ready to take control. Its customers ⁠and partners include Toyota, Mercedes-Benz, SAIC, General Motors, BYD, and Audi.

Designing Charging Hubs for Autonomous Fleets

We are excited to partner with Terawatt for today’s newsletter, one of the leading providers of charging infrastructure. To learn more about Terawatt’s network of AV charging hubs and track record of 99%+ uptime, reach out to Logan Szidik at lszidik@terawattinfrastructure.com

DisplayRide Opens RoadData.AI to Partners: Millions of Miles of Real-World Driving, Structured for Physical AI (link).

RoadData.AI captures continuous, real-world driving across some of the country’s densest and most complex environments, and structures every moment around DisplayRide’s proprietary C³O framework (Cognition, Context, Outcome) — recording not just what happened on the road, but why, and what came of it.

Robotaxis drive miles just to get cleaned and charged; this new startup wants to fix that (link).

Aseon Labs thinks it has a fix: parking space-sized automated pods that can be scattered throughout cities to inspect, clean, and charge robotaxis.

Tesla: Engineering tests of the first production Cybercab have begun in Austin (link). There are just two seats and it has no steering wheel or pedals.

Related: Cybercab spotted on the 405 here in LA. 🤔 (link).

Trump’s Transportation Department Launches Commonsense Updates to Brake Pedal Requirements for AVs (link).

I’ll be speaking at SAE International’s Automated Transportation Symposium on July 30th at 11:40 am on the ‘Journalist Panel’ alongside Jane Lappin, , Alan Ohnsman of Forbes, and David Welch of Bloomberg. Feel free to reach out if you’ll be in attendance!

Lucid overhauls its entire C-suite under new CEO after 18% layoffs (link). Hard to see this as a positive for Uber’s autonomous vehicle partnership with Lucid. Replacing nearly the entire executive team is rarely a sign that everything is going according to plan.

Lucid is replacing almost its entire executive team under new CEO Silvio Napoli, bringing in a new CFO, CTO, and three other C-level leaders in a single announcement.

Tesla: First Responder Information for Robotaxi (link).

What else we're reading/listening to

The Road to Autonomy by Grayson Brulte, featuring Hertz CEO Gil West: Hertz Isn’t Just a Rental Car Company Anymore (link).

Figma Config 2026: Legibility by design ft. Ryan Powell - Waymo (link).

Redefining Travel Time with PIX Moving’s Autonomous Shuttles (link). ’s latest article for The Driverless Digest, where he explore PIX Moving’s vision for autonomous vehicles as “moving spaces”, transforming travel time into opportunities for work, relaxation, tourism, and retail.

Autonomous Charging for Robotaxis with Rocsys CEO Crijn Bouman (link). My latest podcast where we discussed why charging is becoming one of the biggest operational bottlenecks for robotaxi fleets, how the Rocsys M1 hands-free charging system works, what makes an efficient robotaxi depot, and the biggest operational challenges AV fleets will face as they scale, among other things.

AVs/Humans behaving badly

‘Waymo’s pathfinding algorithm is on the struggle bus this weekend’ (link).

Related: ‘Waymo Took Me For Joyride’ (link). Waymo really needs an “Info” button that explains why it’s taking a route that seems out of the ordinary. Even a simple message like “avoiding traffic,” “making a safer turn,” or “road closure ahead” would go a long way toward building rider trust.

‘This (27 min. ETA) is ridiculous, and I have Waymo Premium’ (link). Waymo Premier promises “priority pick-ups”, but the challenge with that benefit is that it’s completely subjective. Unless riders know what the wait would have been without Premier, it’s hard to tell whether they’re actually getting any value from it.

AVs/Human behaving “goodly” :)

‘Because after a night out, getting home should be the easiest part’ (link). I’ve actually been surprised that I haven’t seen or heard many complaints about Waymo’s pickups and drop-offs around the World Cup. Given the crowds and traffic, I expected there to be a lot more friction. Nice job Waymo.

City of Austin: This morning we filled a pothole reported to us by Waymo through Waze for Cities (link). Love this idea for LA, but I don’t think reporting potholes is the problem... it’s actually filling them. 😅

‘Spotify now available on Ojai rides’ (link). But it sounds like it’s still a bit hit or miss. I’m not sure why music integration has been so challenging in the Ojai. I mentioned the same issue in my first ride review, and it still doesn’t seem fully ironed out.

Shout-outs

Big thanks to TDD readers at IATR, Upstream, Hankook, and IIHS-HLDI for referring new subscribers. If there’s someone you think would enjoy TDD, feel free to forward this email or use the referral button below, and we’ll make sure to shout out your company.

Refer a friend

Neat Jobs

Data Scientist, Fleet Operations at Wayve (link) via Agata Rother.

Waymo is preparing for autonomous charging (link), and hiring a Charging Automation Lead (link).

Government Affairs and Public Policy Manager, Cities at Zoox (link) via Phillip Pierce.

Job Moves

Felipe Parodi: Waymo (link).

Andreas Lieber: Serve Robotics (link).

Val Scaglione: Altera -> Zoox (link).

We’re now at 11,000+ subscribers! Interested in advertising with TDD? Check out our media kit here and reach out to us. Newsletter spots are currently sold out until September but we do have podcast advertising opportunities available.

Reach out

Until next week :)

-Harry

Most autonomous vehicle companies are focused on solving transportation.

PIX Moving is looking to redefine what transportation is.

The Chinese autonomous robotics company has spent over a decade developing a radically different vision of driverless vehicles, one where the removal of the steering wheel does more than eliminate the driver. It fundamentally changes the purpose of the vehicle itself.

Instead of treating autonomous vehicles as a more efficient way of getting from A to B, PIX sees them as flexible, human-centred spaces that can adapt to different needs, whether as a shuttle, workspace, café, retail store or even a mobile classroom.

The company calls this concept “moving spaces”.

It is a philosophy rooted in the background of founder Angelo Yu, whose experience spans architecture and engineering rather than traditional automotive design.

“Is mobility only about moving people from A to B, or is it about how people and cities relate to one another?” Yu has said. “I started PIX to rethink mobility not just as transportation, but as a human-city interface.”

Founded in 2013, PIX has grown into a full-stack autonomous mobility company, developing its own vehicles, software and autonomous systems. It now operates approximately 200 autonomous vehicles globally. For comparison, Zoox, which is owned by Amazon, has about 100 active vehicles on the streets.

As PIX expands into Europe, I spoke with Alexandru Vasiliu, Managing Director for Europe, about purpose-driven mobility, autonomous tourism and why travel time may soon matter more than transit itself.

The RoboBus

PIX’s flagship vehicle is the RoboBus, an autonomous shuttle designed for first and last mile and community transportation.

Carrying up to six occupants, with one safety supervisor on board, it travels at speeds of up to 35 km/h (21 mph) and is designed for Level 4 autonomy. There is no steering wheel or driver’s seat. Instead, the interior resembles something closer to a moving lounge than a conventional vehicle.

Passengers sit opposite each other in an airy, panoramic cabin filled with natural light. Large windows maximise visibility while digital displays provide route information, rear camera views and advertising opportunities.

Vasiliu compares the experience to travelling in a gondola.

“When I first rode it in China, I felt like I was on a cable car,” he told The Driverless Digest. “You’re moving through traffic but there’s no stress, no sudden braking, no worries. You simply enjoy the surroundings.”

That experience reflects PIX’s broader question: what happens when the steering wheel is no longer central to vehicle design?

“Level 4 autonomy fundamentally changes the way we think about interiors,” Vasiliu says. “A vehicle can now serve as a shuttle, workspace, logistics platform or hospitality venue. The mission defines the architecture.”

The RoboBus represents one expression of this design philosophy.

Another is RoboShop, an electric, autonomous-ready retail vehicle that can function as a café, pop-up store, exhibition space or cultural venue depending on community needs.

Both vehicles are built on PIX’s modular skateboard platform, allowing multiple form factors to be developed from the same autonomous foundation.

“The future cabin is not a fixed environment,” Vasiliu says. “It is a programmable space.”

Reinventing city exploration

PIX believes tourism could become one of autonomy’s most compelling applications.

In Guiyang, China, PIX runs its largest deployment where a fleet of RoboBuses operate on a “WunderLoop” through the city centre.

Locals have given the vehicles a nickname: the “Potato Bus”.

The WonderLoop service is designed to transform sightseeing into an immersive autonomous experience, transporting visitors around key landmarks on fixed circular routes.

The experience combines mobility, tourism and digital storytelling in a way that feels fundamentally different from traditional public transport.

PIX envisions future deployments where generative AI acts as a personalised guide, automatically delivering information about museums, restaurants and cultural sites as passengers pass them.

Building a European base

While China is PIX’s most mature market, Europe is rapidly becoming a strategic priority.

The company’s most significant European project is at the Politehnica University of Bucharest where the campus effectively functions as a miniature city, complete with multiple lanes, intersections, couriers, kindergartens and public spaces.

“It’s the perfect sandbox for Europe,” says Vasiliu.

PIX is deploying RoboBus services across the site while also establishing what could become one of Europe’s first dedicated autonomous mobility labs, providing students and researchers with access to its autonomous technology.

“We want to create the first laboratory for autonomous studies in Eastern Europe,” Vasiliu says.

Germany is another key focus, with public demonstrations in Chemnitz transporting officials as part of ongoing homologation and pilot programmes.

Looking ahead

PIX aims to double its deployed fleet from 200 to 400 vehicles by year-end.

Europe is expected to play a major role in that growth, with multiple pilot projects and commercial deployments currently under development.

The broader question PIX poses is an intriguing one. If travel time no longer requires our attention, what should we do with it instead?

PIX believes the answer lies not in transportation, but in experience.

If the company succeeds, future vehicles may look less like cars and more like programmable spaces designed around how people want to live.

Listen to this episode on Apple Podcasts, YouTube, Spotify, and Substack. Transcripts are also available by clicking on the “transcripts” button in the top right corner of this post on the app/web version. You can also access them from the email version by clicking here.

In today’s episode, I’m speaking with Crijn Bouman, CEO and Co-founder of Rocsys. Crijn shares his background in the EV charging industry, what led him to found Rocsys, and why he believes charging has become one of the biggest operational bottlenecks for autonomous vehicle fleets. We discuss why depot operations deserve far more attention as robotaxi fleets scale, how much of fleet management can ultimately be automated, and why charging is the logical place to start.

Crijn also discusses the Rocsys M1, the company’s hands-free charging solution designed specifically for autonomous fleets. We explore how the system works, why it is designed to be vehicle- and charger-agnostic, the challenges Rocsys faced bringing it to market, and what operators should consider when designing an efficient robotaxi depot. We also cover where the M1 is already deployed, what the AV depot of the future could look like, and the common mistakes companies make when planning fleet operations.

Finally, we zoom out to discuss the global robotaxi landscape, including the differences between the U.S. and Europe, what Europe needs to do to become more competitive, and Crijn’s vision for how automation will transform depot operations over the coming years.

Chapters

• (00:00) Introduction to Crijn Bouman

• (01:30) Crijn’s background, his history in the charging industry, and how he founded Rocsys

• (04:31) How Rocsys supports AV fleets

• (05:30) How much of AV fleet management can be automated?

• (07:35) Why charging is such an important part of AV depot management, and why it should be automated first

• (09:51) Initial challenges faced with automating AV fleet charging

• (13:22) The Rocsys M1 hands-free charging solution for AV fleets

• (14:50) The ease of deploying the Rocsys M1, and its other benefits

• (18:20) Where the Rocsys M1 chargers are currently deployed and future deployments

• (20:15) The ideal robotaxi partner/depot layout for deploying the Rocsys M1

• (22:40) What the AV depot of the future will look like

• (27:05) Common oversights AV companies make when designing depot operations

• (28:28) Rocsys’ global scale and the benefits of operating from the U.S. and Europe

• (30:00) What Europe needs to do to catch up in the robotaxi space

• (32:10) How Rocsys will transform the AV depot of the future

• (35:15) What can’t be automated in the AV fleet depot, and why

• (36:47) Conclusions and final thoughts

Notes/Links:

• You can find Crijn Bouman on LinkedIn and X.

• Rocsys website (link).

• Check out the Rocsys M1 (link).

-Harry

Top Stories of the Week

Amazon’s Zoox unveils redesigned robotaxi ahead of upcoming expansion (link). Funny how everyone calls the Zoox vehicle a “toaster” now. Jokes aside though, it’s cool to see them iterating on the design so quickly. One advantage of building a purpose-built vehicle is that you can make these kinds of refinements without waiting years for a traditional model refresh. Looking forward to seeing Zoox launch paid rides later this year.

The company said it’s equipping the vehicles with higher-quality touchscreens, more comfortable seats and headrests, and small interior tweaks that will make it easier for passengers to spot forgotten items like keys and phones. Zoox is also enlarging and relocating the robotaxi’s “bidirectional reflectors,” which help riders and others such as law enforcement distinguish the vehicle’s front from its rear, so that they’re easier to spot.

Lyft CEO: we’re setting a multi-sensor safety standard for autonomous rides (link, no paywall). Interesting stance. Tesla was the first company I thought of when I read this, given they are the only major AV developer pursuing a camera-only approach. But this op-ed also highlights how platforms like Lyft are establishing their own technical and safety standards for AV partners, rather than simply accepting whatever technology an operator chooses. That said, I’m not sure I agree with evaluating AVs this way. As we’ve written before, I’d rather see an outcome-based framework that judges systems on real-world safety performance, not the specific technology or sensor stack they use.

As we’ve evaluated various systems, we’ve come to a conclusion: To meet our safety standards, autonomous vehicles must have a multi-sensor approach before we allow them on the Lyft platform. To be clear, this policy applies to fully driverless vehicles operating on the Lyft platform, not to driver-assistance features used by human drivers.

Related: Lyft’s Hybrid Approach to Autonomy with CEO David Risher (link).

Excited to announce that we’ve brought on Ben Hubbard, formerly of Zag Daily, as our new Editor on The Driverless Digest! (link).

Other Stuff

AI is coming for truck drivers. A new bill is trying to brace US workers for impact (link, no paywall).

New global rules clear the road for driverless vehicles (link).

Related: UK government drafts 10 safety principles for autonomous vehicles (link).

Rocsys M1 Keeps Robotaxi Fleets Moving. Bay by Bay.

This edition of The Driverless Digest is brought to you by Rocsys. Their M1 system, an overhead rail-mounted solution, integrates with virtually any depot layout and can charge multiple vehicles across up to 10 bays, supporting safe, uninterrupted service. Compatible with any charger and vehicle type across mixed fleets, the M1 enables AVs to get back on the road faster while improving the bottom line. To learn more, reach out to Erin Galiger, Rocsys Director of North American Markets, at egaliger@rocsys.com.

Discover Rocsys

Waymo Charging Partner Terawatt Raises Up to $300 Million in Debt (link, no paywall).

The funding underscores investor confidence in the rollout of robotaxis as operators like Waymo seek to offer more commercial rides in the US and beyond.

Related: Charging the Fleets of the Future with Terawatt CEO Neha Palmer (link).

TaskUs Survey: Robotaxi Adoption Continues to Surge, with 38% of Americans Likely to Try One in the Next 12 Months (link).

Consumers’ top two sentiments surrounding robotaxis are curiosity (42%) and nervousness (41%). 41% of consumers haven’t ridden in a robotaxi because they haven’t had the opportunity. Nearly one-quarter (23%) of consumers who live in a city offering robotaxis fully trust them, compared to 4% of consumers who don’t. 38% of Americans are likely to try a robotaxi in the next 12 months.

Waymo continues expansion of robotaxis in Nashville (link).

Nashville riders previously had to sign up on a waiting list before they could hail a robotaxi. But on Thursday the company announced anyone in the city's service area could sign up for the Waymo app and ride instantly without a waiting period.

Komatsu becomes first OEM to commission 1,000 ultra-class autonomous haul trucks (link).

‘At this point, I’m not sure there are any human drivers on the road in LA after midnight’ (link).

‘Cambridge Electric Transport and Osmosis AI have signed an MoU to collaborate on autonomy trials using the CitiPod’ (link).

CaoCao and May Mobility Announce Strategic Partnership to Pioneer Robotaxi Expansion in Europe (link).

‘New Waymo Hyundai spotted in DTLA’ (link).

Waymo Registers German Entity for Future European Expansion (link, no paywall).

The company registered a local entity called Waymo Germany GmbH that will “offer ride-hailing services with autonomous vehicles and provide services that support the commercial offering of such services by third parties,” according to a company registration filing.

‘Go eyes robotaxis and acquisitions after Japan’s biggest IPO of 2026 — here’s why it matters’ (link).

Go, which raised ¥88.6 billion ($553 million) in its IPO on Tuesday, plans to allocate ¥8 billion ($50 million) of the proceeds to expanding its robotaxi business and pursuing acquisitions, according to a company spokesperson.

‘Can you import autonomous driving?... This is what running AVs actually looks like from the inside.’ (link).

"Wheels up!" by Southwest Airlines was a great slogan, emphasizing that value is created only when flying. The same dynamic is true in the robotaxi market: money is earned on the road, and at the current stage where commercial viability is the next milestone, and roborides are a fraction of the total ridehail demand you just got to get your vehicle out there as frequently as possible” (link).

Rivian owners file lawsuit alleging false promises on self-driving features (link). Can’t say I blame them.

The class-action complaint, which was filed Wednesday in the U.S. District Court for the Central District of California, focuses on the first-generation models of the R1T and R1S and claims that Rivian represented that these flagship vehicles would be capable of hands-free, eyes-off driving.

Waymo Has Been Defeated by New York City (link).

WeRide, Geely Farizon and Kwoon Chung launch RHD robotaxi (link).

Tesla claims driver ‘manually overrode self-driving’ in deadly Texas crash (link, no paywall). This theory makes sense, since Tesla’s FSD Supervised generally isn’t going to drive 70+ mph through a residential neighborhood on its own. It’s a tragic crash, but based on what’s been reported so far, I don’t really see how Tesla should be at fault here.

Tesla is pushing back on claims that its Full Self-Driving (FSD) system caused a fatal Texas crash, where a speeding Model 3 barreled into a home, killing a 76-year-old woman inside. In a reply on X, Tesla AI head Ashok Elluswamy says the driver “manually overrode self-driving by pressing the accelerator all the way to 100%.”

‘Waymo World Cup traffic outside SoFi stadium for the US vs Turkey match’ (link).

What else we’re reading/listening to

My First Ride in Waymo’s New Ojai (link).

Inside The Ride: Scaling Zoox | Episode One (link). It was cool to hear how much passion Jesse had in this interview and all of the technical details he shared. Excited to hear his keynote at ATS 2026, where I’ll also be speaking on the Journalist Panel on Thursday, July 30, at 11:40 am. Drop me a note if you’ll be in attendance, there’s an impressive lineup of speakers.

INTERVIEW With Pony AI CEO James Peng: “In two years, we’ll be bigger than Waymo on fleet size” (link).

The Last 50 Ft by Autolane: Ep #12 | The Rideshare Guy Got in a Waymo. Everything Changed (link).

We’ve released recordings from the Detroit Urban Autonomy Summit. The audio quality wasn’t amazing but all the more reason to attend our next one in person in Austin on 10/14.

• A New Vision: Improved Technology for Safer Streets & Highways (link).

• Charged Up Infrastructure: Electrification’s Capital Moment (link).

• Owned vs On-Demand: Which Way for AVs? (link).

AVs/Humans behaving badly

Teens seen hanging out of Waymo in busy Santa Monica traffic (link). Honestly, this sounds exactly like something that teenage boys would do.

‘Waymo driving in oncoming traffic’ (link).

‘Zoox - what do u mean call failed’ (link). Not quite sure what happened here, but it did make me wonder how you’d exit a Zoox in an emergency1. With Waymo, you can simply pull the door handle and get out at any time.

‘Damage to Waymo Seat’ (link).

‘Waymo robotaxi stuck in high water on Fitzhugh Avenue’ (link).

AVs/Human behaving “goodly” :)

‘My Cybertruck, on full FSD, in downtown Los Angeles, suddenly swerved to the left and narrowly avoided a collision with the white Audi car changing lanes very fast. If I were the driver, I wouldn’t have been able to prevent the accident’ (link). Impressive! Not sure how you miss a Cybertruck though 😂.

Shout-outs

Big thanks to TDD readers at MOIA, Vay, and to Robert V for referring new subscribers. If there’s someone you think would enjoy TDD, feel free to forward this email or use the referral button below, and we’ll make sure to shout out your company.

Refer a friend

Neat Jobs

Product Manager, Sensing Resilience at Waymo (link) via Volker Grabe.

Policy & Regulatory Affairs Manager at Zoox (link) via Jonathan Changus.

Quality Management Lead at Wayve (link) via Tim Dawkins.

Multiple Mobility/AV/FleetOps roles at Moove, via Arun Victor

Multiple autonomy roles at Rivian, via Mikael Thor.

Job Moves

William Joel Sánchez: Aurora -> Nuro (link).

Taylor Powell: Carnegie Robotics -> Waymo (link).

Sami Pukkila: Polestar -> Wayve (link).

Alireza Abbaspour: Qualcomm -> Waymo (link).

Lando Juarez: Uber -> Bedrock Robotics (link).

We’re now at 11,000+ subscribers! Interested in advertising with TDD? Check out our media kit here and reach out to us. Newsletter spots are currently sold out until September but we do have podcast advertising opportunities available.

Reach out

Until next week :)

-Harry

Some eagle-eyed readers may have spotted a new byline in our recent interview with Pony AI’s CEO: meet Ben Hubbard.

Ben joins us from Zag Daily, where he served as Executive Editor and helped build one of the most respected publications covering the future of transportation. Ben has been covering the sustainable mobility sector for six years with his reporting spanning autonomous vehicles, delivery robots, drones, eVTOLs and the wider mobility ecosystem.

Ben is deeply curious about how autonomous tech is moving from promise to reality. He has a knack for uncovering hidden stories and identifying the trends that matter, and we’re excited to have him leading editorial coverage at The Driverless Digest.

Talk to Ben!

One of Ben’s priorities is getting to know The Driverless Digest community.

He’s opened up time in his calendar this week for conversations with our readers. Whether you have a story tip, an interesting perspective on the AV market, a partnership idea, feedback on the publication, or just want to introduce yourself, he’d love to hear from you.

If you’d like to connect, book a time with Ben here: [link]

Welcome, Ben.

I recently took my first ride in Waymo's newest vehicle, the Ojai. Overall, I came away impressed with the passenger experience, though there are still a few areas where I think there's room for improvement.

Calling For a Ride

Since the Ojai vehicles are currently only available to Waymo’s Trusted Tester program, I had to enlist the help of a friend to request the ride.

Dusty has been a Trusted Tester in Los Angeles for a while and joked that he hadn’t seen much benefit from the program until the Ojai rollout.

When requesting a ride, he had the option to choose between “First Available”, the traditional Jaguar I-PACE fleet, or the new Ojai vehicle. Interestingly, I’ve heard that some Trusted Testers only see the ‘First Available’ option and aren’t able to specifically request an Ojai vehicle. Lucky Dusty.

Availability also seemed quite good. Our vehicle was only about four minutes away on the first ride (and five minutes away on the second), and during non-peak hours there appeared to be plenty of vehicles on the road.

The other big benefit for Trusted Testers is that all Ojai rides are free!

First Impressions

The first thing that stood out to me was how spacious the vehicle felt upon entry. Compared to the Jaguar I-PACE that Waymo has used for years, the cabin feels more open and purpose-built for passengers. And for someone like me who stands at 6’ 3”, there’s a ton of leg room in the back seats and front seat.

The back seats didn’t recline and were a little too upright for my liking after 20-30 minutes of driving (I have a tight lower back) but the front seat had all the bells and whistles and I was able to adjust the seat to my liking.

I loved the sliding doors which allow you to enter and exit from either side of the vehicle. It seems like a small thing, but it makes pickups and drop-offs more convenient and gives the vehicle a futuristic feel. It also reduces the risk of dooring cyclists or other cars, which is a nice safety benefit.

The overall fit and finish were also impressive, not as high quality as a luxury BMW or Mercedes, but definitely solid. The seats were comfortable, there was plenty of room to stretch out, and the vehicle offered many of the creature comforts you’d expect from a premium ride. Wireless charging, cup holders, three very high quality large infotainment screens, and other passenger-focused features help reinforce that this is a vehicle designed around the rider experience. I did notice that it was missing a glass sunroof though, like the I-PACE and many other newer EVs.

There’s also ample room in the back for luggage.

What I Didn't Love

While the passenger experience was excellent, I found the driving style to be a bit less polished than what I’ve experienced in Waymo’s Jaguar I-PACE fleet.

The vehicle wasn’t bad by any means, but it felt slightly jerkier at times and not quite as smooth as the I-PACE. That said, it still felt totally safe throughout the ride. The best way I can describe it is that the Ojai feels like it knows how to drive but still has its training wheels on. The fundamentals are all there, but some of the confidence and refinement haven’t fully arrived yet.

I also noticed the vehicle being a bit more cautious than necessary in a few situations. At one point, it slowed down and changed lanes to move around a jogger who wasn’t really in its path (they were in a painted bike path). While I’d always rather see an autonomous vehicle err on the side of caution, the maneuver felt more conservative than necessary.

The 3 Infotainment Screens Are Great!

The three infotainment screens are high quality and do an excellent job showing surrounding vehicles in detail, and overall the in-cabin tech feels thoughtfully designed for passengers.

That said, the audio experience could use some improvement. The speakers were adequate, but they didn't feel particularly premium, especially compared to what many riders are accustomed to in modern vehicles. I was also surprised that Spotify integration isn't available yet, especially as robotaxis increasingly compete with personally owned vehicles and premium ride-hailing services where entertainment and personalization matter more.

Overall

Despite a few nitpicks, I came away impressed. The vehicle feels thoughtfully designed, comfortable, and purpose-built for autonomous transportation. The passenger experience is arguably the best I’ve seen in a robotaxi so far.

The driving style still feels a bit conservative and could use some refinement, but if this is where the platform is today, it’s easy to see how quickly it could improve as Waymo continues to iterate.

- Harry

Many people likely hadn’t heard of Pony.ai until March this year, when it was one of three companies behind Europe’s first commercial robotaxi launch in Zagreb.

Recently, Pony announced it was running autonomous vehicle testing in Luxembourg with Bolt and Stellantis.

At MOVE London 2026, I sat down with CEO James Peng to explore how Pony is approaching Europe’s fragmented autonomy landscape city by city.

What is Pony?

Pony AI Inc is a public autonomous driving technology company founded in Silicon Valley, but now firmly rooted in China, where it runs fully driverless Level 4 commercial services in Beijing, Guangzhou, Shenzhen and Shanghai.

Outside China, it is present in Singapore, Doha, Dubai, Seoul, Luxembourg and Zagreb.

Pony’s main role is to provide the “Virtual Driver”, a vehicle-agnostic autonomous driving system that can be deployed across robotaxis, robotrucks, and low-speed urban services such as delivery and maintenance vehicles.

Pony currently has a fleet of nearly 2,000 Gen-7 robotaxis.

“In two years, we’ll be bigger than Waymo on fleet size,” says Peng.

“At the start of 2025 we had 270 robotaxis, today we’re at 2,000 and by the end of this year we’ll have 3,500, so we’re on a strong growth trajectory.”

The company delivered its first profitable quarter in Q4 last year. In Q1 this year, it generated $8.6 million in robotaxi revenue, up 395% year-on-year.

Pony is now valued at roughly $3.6 billion.

European strategy

Pony’s Europe strategy is deliberately conservative and partnership-led, avoiding a full stack operating model.

“The European market, first and foremost, is a very interesting market to us,” Peng told The Driverless Digest. “It has a population of 500 million people and is economically very strong.

“But we are definitely not going to come to Europe alone. We will work closely with our partners.”

Uber and Bolt sit at the center of that approach, alongside city specific fleet operators depending on the market.

Early rollouts focus on tightly defined populous urban zones where demand is concentrated and utilization can be maximized. James says it is important for Pony to prioritize density over wider coverage in the early stages to prevent long wait times and achieve more rides per vehicle.

“We’re getting 25 rides per vehicle per day with an average wait time of five minutes in Shenzhen and that’s what we’re looking to bring to Europe,” Peng adds.

For context, Waymo’s most recently reported California data showed its San Francisco fleet averaging roughly 24 rides per vehicle per day as of December 2024. While the two figures are not perfectly comparable given differences in service areas, fleet composition and operating conditions, they suggest Pony is already achieving utilization rates comparable to those of the industry leader in at least some markets.

Luxembourg and Croatia

The early shape of that strategy is already visible in Croatia and Luxembourg.

In Zagreb, Pony works with Uber on the ride-hailing side, while Verne handles local operations. Around 15 robotaxis are currently running through the downtown area and airport routes, with a safety operator onboard.

“Our internal data suggests it is safe to remove the safety driver, and we will be testing that very soon,” Peng says.

In Luxembourg, Pony is partnering with Bolt and Stellantis to launch a testing programme through which it can validate safety, operational performance and regulatory readiness for its robotaxis.

“With Luxembourg, you can make an appointment with a minister very quickly. For us, that is great because it means we can grow fast while working with policymakers to demonstrate to the public that the technology is safe, reliable and convenient.”

Challenges in Europe

Peng says the two key factors shaping European deployment are regulation and partnerships.

“We must have strong partnerships with automakers, ride-hailing platforms and service providers to scale.”

Regulation remains the biggest structural challenge. Autonomous vehicle developers continue to face fragmented testing permits, approval processes, road rules and data standards across different countries and cities.

“We are already seeing much more discussion at the policy level and many more cities have a waiting list for autonomous vehicle trials,” he adds.

On 9 June, Europe moved a step closer to harmonising this landscape, with 17 countries agreeing to coordinate autonomous vehicle testing across borders.

2026 outlook

Peng confirms that another European city is already “cooking”, although details remain under wraps.

“There will be something this year,” he says. “There is a lot happening.”

The plan is to be in 20 cities globally by the end of this year.

For now, the strategy remains consistent. Start small, build partnerships, demonstrate safety and then scale.

Top Stories of the Week

Self-driving tech supplier Mobileye targets U.S. robotaxi launch in 2027 (link).

Mobileye said it would launch its own robotaxi service, planning to deploy an initial 100-vehicle fleet into a major U.S. city in 2027. The company said the first launch will be phased throughout next year, after which they’ll aim to scale the fleet to roughly 17,000 vehicles over the next five years.

Other Stuff

Element partners with Waymo to enable the future of autonomous mobility at scale (link).

This partnership, enabled through Element Mobility, will provide end-to-end fleet management and operational services to support Waymo's autonomous mobility operation, beginning with an initial deployment in San Diego and expanding to additional markets over time.

Uber is bringing robotaxi service to Houston (link).

Houston will be the second market for the robotaxi service after a launch in the San Francisco Bay Area later this year. Service is expected to expand to dozens of additional markets in the coming years. Uber has secured a 50,000-square-foot depot facility in downtown Houston and a dedicated charging pitstop in the city, with construction expected to begin in early 2027.

Robotaxi rankings show Waymo lead, China’s rise (link). This is a neat idea, but I’d be curious to know what underlying data and methodology were used to rank each company. Otherwise, it’s hard to know what the rankings actually mean.

Uber and WeRide to launch robotaxis in Zurich (link). I’m not sure how meaningful these partnerships are since the biggest barrier to AV deployment in Europe seems to be regulation, not technology.

Rocsys’ Charging Solution Sets the Pace at the Depot

We are proud to partner with Rocsys for this week’s edition of The Driverless Digest newsletter. Their M1 is the first overhead rail-mounted hands-free charging solution for robotaxi fleets, automating the charging process across multiple bays. The system enables uninterrupted operations and up to $1.7 mil. in annual savings for a 50-bay depot. To learn more, contact Erin Galiger, Rocsys Director of North American Markets, at egaliger@rocsys.com.

Discover Rocsys

Waymo recalls nearly 4,000 robotaxis to stop them driving into highway construction zones (link).

This is the sixth recall Waymo has issued for its robotaxis. In May, the company recalled its robotaxis after they drove into flooded roads, and in December, it issued one to address its vehicles’ illegal behavior around school buses. Waymo has previously issued recalls to fix low-speed collisions with chains and gates and telephone poles, and one to solve a problem regarding towed trucks.

Cathie Wood Asks Elon Musk How Tesla Handles ‘Parking And Traffic Violations’ After The Robotaxi She Rode Gets $75 Ticket (link). Nice cameo from the Zoox mapping vehicle at ~7:30 too 🤣

Rivian’s CEO on Tesla’s Cybertruck, Ferrari’s Luce, and What Happens If the R2 Fails (link, no paywall). This was a great interview by WIRED’s Jeremy White that covered a lot of topics you don’t typically hear discussed. I especially appreciated that he asked some tougher questions around competition, product strategy, and execution risk rather than sticking to the usual CEO softballs.

Rivian CEO RJ Scaringe: “In a world where there’s a very small number of people that develop autonomous capabilities, it will likely still carry some charge or premium. Once enough companies have developed it, what you can charge will become much lower. But in either scenario it’s very important. Where it becomes free, then it becomes a necessary component to sell vehicles.”

Related: ‘First Drive with Universal Hands Free (UHF) on the Rivian R2!’ (link). Personally, I love my Rivian (link) R1S but UHF feels like an early version of Tesla Auto Pilot. I also liked Omar’s quote in the video, “Overall, it’s ok. It’s better than nothing”. Well said.

Tesla ‘Robotaxis’ are not crashing because they are not running (link).

A program with 14 unsupervised vehicles is going to have very few crashes by definition. That’s not a vindication of the technology; it’s the statistical consequence of barely operating. The real test only comes when Tesla puts hundreds or thousands of driverless cars on the road and the crash count is measured against actual rides delivered — the way it is for Waymo.

‘City #5 is officially live! Avride is expanding its partnership with UberEats to launch autonomous robot delivery in Arlington, Virginia’ (link).

Autolane: Over the past 40 weeks, our autonomous vehicles helped complete more than 1,000 meal deliveries throughout downtown San Francisco, helping seniors and those in need receive the meals they depend on (link). 👏

We’re also excited to welcome Autolane as our latest partner for the Austin Urban Autonomy Summit, presented by Nexar, taking place on October 14th. Reach out if you’d like to learn more about a partnership.

The Cybercab is the lightest, most efficient Tesla ever made (link, no paywall).

TaskUs Forward ‘26 is coming up on October 28, 2026, at Above the Rim at Chase Center in San Francisco, CA. Sessions will focus on breakthroughs in agentic AI, robotics, autonomous vehicles, and more (link).

Waymo curbside drop-off now available at San José Mineta International Airport (link). Great, now I just need a good reason to go to San Jose.

GM killed Cruise, but don’t count it out of the robotaxi race (link, no paywall).

Anderson said GM's approach is to develop self-driving technology by breaking the driving experience into pieces and examining where autonomy is most useful to car owners. That means first tackling long stretches of highway driving before expanding to arterial roads and urban centers. Over time, the executive said GM's autonomous driving systems will be able to operate in enough regions to make a viable robotaxi service.

‘Spotted in SF! New Waymo wrap, and she is SO cute!!’ (link).

Rivian CEO says company will release tech similar to Tesla FSD later this year (link, no paywall).

Later this year, we'll have full supervised point-to-point, which will be very similar to Tesla's FSD," he said. "And that'll roll out to all of our Gen 2 vehicles and, of course, R2.

What else we're reading/listening to

Just ordered ’s new book, and excited to dig in. The book explores the public debate surrounding Tesla’s driver-assistance systems and the sweeping claims that have shaped the company’s place in technology, culture, and the marketplace (link).

The Next Mile by : Are Self-Driving Cars Safer than Public Transit? (link).

Inside Waymo’s Interview Process: An Engineer’s First-Hand Experience (link). Our latest podcast where we discussed Waymo’s interview process from a candidate’s perspective, how the company evaluates engineering talent, and what aspiring AV engineers can do to stand out in one of the industry’s most competitive hiring pipelines, among other things.

AVs/Humans behaving badly

‘Autonomous Uber crash’ (link). You could argue the AV should have done a better job avoiding this crash, but the safety driver appears to have done an excellent job and was already preparing to take over before impact.

‘Do better’: San Francisco is fed up with this Waymo behavior (link, no paywall). Uber and Lyft drivers do this all the time, but I’m not sure that makes it acceptable for Waymo. Cities should enforce these rules, Waymo should avoid this behavior, and municipalities should probably create more dedicated pickup and dropoff space too.

AVs/Human behaving “goodly” :)

Interesting prompt I saw for the first time today in a Waymo. Would be smart to develop a rating score for previous passengers (link).

Waymo: Because we react to the unexpected in milliseconds (link). Amazing work by the Waymo here. But two people riding on one scooter downhill at night with no helmets? That’s a Darwin Award contender right there.

Shout-outs

Big thanks to TDD readers at Uber, and to Simon C for referring new subscribers. If there’s someone you think would enjoy TDD, feel free to forward this email or use the referral button below, and we’ll make sure to shout out your company.

Refer a friend

Neat Jobs

Field Application Engineer at Rocsys (link) via Rocsys.

Multiple Business Partner roles at Waymo, via Sheila Nemeth.

Senior/Staff Technical Program Manager at Zoox (link) via Taylor Arnicar.

Job Moves

Yuezhan Tao: UPenn -> Zoox (link).

Paul Reynolds: Oxa -> TaskUs (link).

Court C: ASARCO -> Zoox (link).

We’re now at 11,000+ subscribers! Interested in advertising with TDD? Check out our media kit here and reach out to us. Newsletter spots are currently sold out until September but we do have podcast advertising opportunities available.

Reach out

Upcoming Event: Our next Urban Autonomy Summit will be taking place on October 14th in Austin! You can check out some photos from our Detroit edition here.

Until next week :)

-Harry

Listen to this episode on Apple Podcasts, YouTube, Spotify, and Substack. Transcripts are also available by clicking on the “transcripts” button in the top right corner of this post on the app/web version. You can also access them from the email version by clicking here.

In today’s episode, I’m speaking with an anonymous AI and machine learning engineer with a background in computer vision and autonomous systems. After advancing to the final round of interviews for a Senior Software Engineering role at Waymo, he joins me to share a rare candidate’s perspective on one of the most sought-after engineering jobs in autonomy.

We discuss Waymo’s interview process from start to finish, including the types of technical questions candidates can expect, how the company evaluates engineering talent, and what stood out compared to interviews at other autonomous vehicle companies such as Zoox where he has previously been interviewed at. He shares insights into Waymo’s engineering culture, the skills and experience the company appears to value most, and why preparing for an autonomy-focused role can be very different from a traditional software engineering interview.

The conversation also explores broader hiring trends across the AV industry, what it’s like interviewing during a period of innovation, rapid growth and commercialization, and how candidates can position themselves for success at leading AV companies. Finally, he reflects on what surprised him most throughout the process and what aspiring engineers should know before pursuing a role in the industry.

Chapters

• (00:00) Introduction to our guest

• (03:00) What made him want to work for Waymo

• (04:30) How the Waymo user experience influences potential employees

• (05:31) How job applicants perceive working at Waymo

• (06:50) The Waymo application and interview process

• (11:20) Tips for the Waymo online job application

• (14:12) Feedback during the interview process and screening tips

• (15:44) How Waymo helps applicants prepare for the interview process

• (16:20) How he prepared for the interview and the role of AI in his interview prep

• (18:40) How Waymo’s interview questions differ from those at other companies

• (21:57) Something he wished he knew before the interview process

• (22:48) The Waymo vs. Zoox interview process

• (24:50) His reaction to being rejected for the role

• (28:05) What Waymo applicants should do to stand out from their peers

• (31:34) How to position yourself for a future role in an AV company

• (34:22) His future plans for working in the AV space

• (36:05) Conclusions and final thoughts

Designing Charging Hubs for Autonomous Fleets

We are excited to partner with Terawatt, one of the leading providers of charging infrastructure. To learn more about Terawatt’s network of AV charging hubs and track record of 99%+ uptime, reach out to Logan Szidik at lszidik@terawattinfrastructure.com

Notes/Links:

• Waymo’s careers page (link).

• Zoox’s careers page (link).

-Harry

Top Stories of the Week

Waymo introduces $30-a-month premium tier for riders who want faster pickups (link, no paywall).

Riders will receive a number of perks, including priority pickups, 10 percent cash back on every trip, early robotaxi access in new cities, and up to five free cancellations a month. Waymo Premier will be initially offered to select riders in San Francisco, Los Angeles, and Phoenix.

I like it. Priority pickup will probably be the biggest benefit for riders. And on the business side, this is a smart way to segment customers. Waymo has fewer levers than Uber or Lyft to manage supply and demand, since it can’t just use driver incentives to bring more cars onto the road during peak demand times. So it makes sense to prioritize your most loyal riders while creating a new recurring revenue stream in the process.

Urban Autonomy Summit

We had a great turnout at the Detroit edition of our Urban Autonomy Summit, presented by Nexar, with over 200 attendees! It was exciting to bring the event to the cradle of the U.S. auto industry and connect with so many people shaping the future of mobility and autonomy. Big thanks to all the TDD readers who attended, and to our partners at Terawatt, TaskUs, Rocsys, and Voltera for helping make it happen.

We're heading to Austin next (link) on October 14, followed by San Francisco on January 27, 2027. Reach out if you’d like to get involved as a partner. Spots are filling up fast!

Other Stuff

Bolt, Stellantis, Pony.ai to launch autonomous vehicle test program in Luxembourg (link, no paywall).

Their pilot program ​will focus on validating the safety, ​performance and regulatory readiness of Pony.ai’s autonomous ‌vehicles ⁠in Luxembourg’s traffic environment, the companies said. Stellantis will contribute a mid-size van based on its L4-Ready platform, they ​added

Maybe I’m becoming harder to impress, but it’s difficult to get too excited about another AV pilot with a safety driver behind the wheel. Europe has no shortage of autonomous vehicle announcements, pilots, and demonstrations, but the question I keep coming back to is: where are the driverless rides?

And to be clear, this isn’t a knock on Pony.ai’s technology. We already know Pony can operate driverless services in China. The more interesting question is what Europe needs to do from a regulatory and operational standpoint to move beyond safety-driver pilots and toward actual driverless deployments.

At this point, I’m almost more impressed by companies like Bliq pursuing an advanced L2/L3 approach with remote supervision and no one behind the wheel than an “L4” pilot that still relies on a safety driver. If a human is sitting there ready to take over, it’s hard to know how autonomous the system really is. The real milestone isn’t launching another pilot. It’s removing the driver.

Related: Europe’s First Driverless Vehicle Approval with Bliq CEO Julian Glaab (link).

‘Europe does not have an AV pilot problem. Europe has an AV commercialization problem’ (link).

Lyft Flexdrive CEO, John Parks: Big milestone for the Flexdrive team this week as our staff has fully taken over day-to-day fleet operations for Waymo out of their temporary depot in Nashville (link).

Terawatt’s energy mission (link).

Pittsburgh firm wants parking fees for driverless cars (link). Interesting piece on Meter Feeder that mentions our coverage of Waymo’s snooze fee feature. Chime in on the discussion here.

Welcome to the Waymo World Cup (link, no paywall).

Match attendees can catch driverless rides to six of the 16 North American venues: stadiums in Atlanta, Houston, Los Angeles, Miami, and the San Francisco Bay Area.

Uber Opens London Waitlist for Wayve Robotaxis Ahead of Launch (link, no paywall).

Customers in the UK can join an “interest list” in the Uber app starting Monday to indicate their preference and increase their chances of being matched with a Wayve robotaxi when the service begins, Uber said in a statement. The ride-hailing company didn’t disclose a timeline for the launch.

Another European robotaxi pilot with safety drivers. But to Wayve’s credit, they’re at least pursuing a mapless approach, which gives them a compelling scalability story. If Wayve can prove its technology works across cities without extensive mapping, expanding into new markets could be much faster and less capital intensive than traditional robotaxi approaches. The real challenge, though, isn’t launching a pilot. It’s removing the safety driver and scaling commercially.

Streamlining Charging Operations, Hands-Free and At Scale

We are excited to have Rocsys, the global leader in hands-free charging solutions, as the sponsor this month. Rocsys enables robotaxi fleets to maximize uptime and throughout by delivering consistent, predictable performance that allows vehicles to charge seamlessly, reliably, and safely as soon as they return to the depot. To learn more, contact Erin Galiger, Rocsys Director of North American Markets, at egaliger@rocsys.com.

Discover Rocsys

Motional: Cracking the Long-Tail Code in Autonomous Driving with nuReasoning (link).

‘The future of ride-hailing in Southeast Asia will not be fully autonomous’ (link).

‘Letters: These riders can’t use Waymo’s driverless cars and are treated like second-class customers’ (link, no paywall).

Waymo’s lack of wheelchair accessibility has generated criticism from some disability advocates, but I think the issue is more nuanced than it’s often portrayed.

Uber and Lyft drivers are notorious for issues with riders traveling with service animals, so Waymo has arguably improved accessibility for many riders. And the company’s new Zeekr-based vehicle includes accessibility features like braille labels and audio guidance.

But Waymo still doesn’t have a great solution for riders who rely on wheelchairs and cannot transfer into a standard vehicle on their own. Today, those riders are generally directed to a human-operated wheelchair-accessible vehicle service rather than an autonomous vehicle, which makes sense since the whole point of Waymo is that there is no driver.

At the same time, the technology to support wheelchair users isn’t new. All licensed London black cabs are wheelchair accessible, demonstrating that accessible vehicle fleets can work at scale. The bigger question is whether accessibility should be built into the first generation of the Ojai or whether it’s reasonable to expect companies like Waymo to address it in future vehicle platforms. Personally, I’m sympathetic to both sides. Ojai is still an early product, but if robotaxis are going to become a meaningful part of the transportation system, wheelchair accessibility can’t remain an afterthought forever, especially since Waymo has rightly made accessibility and mobility independence a core part of its broader mission and messaging.

Tesla’s Robotaxi Falls Short With Long Waits and Stalled Rides (link, no paywall).

Almost a year later, Tesla has just 59 vehicles in its entire robotaxi fleet as of Tuesday, limited to three Texas cities.

Ojai is live! (link). I think it’s interesting that the Ojai and the I-PACE are two separate options on the Waymo app even though they both hold four passengers. The downside of this setup is that you’re effectively bifurcating the fleet, which could lead to longer ETAs since Waymo now has to match riders with a specific vehicle type instead of drawing from a single pooled fleet.

‘Nuro test vehicle in SF.’ (link).

Waymo bought Apple’s self-driving car proving ground for $220M (link). The sale was recorded in June 5, and the massive 5,500-acre proving ground includes a 115-acre city course, a 35-acre vehicle dynamics area, a four-mile oval track, and a freeway course purpose-built for autonomous vehicle testing.

PepsiCo and Gatik Announce Multi-Year Agreement to Deploy Autonomous Freight in North America (link).

Waymo says it built a better benchmark for comparing robotaxis to humans (link).

The biggest difference between this new model — which Waymo calls the Reference Driver — and its predecessor is that it is able to reproduce a human driver’s behavior in the run-up to a crash. Previously, Waymo’s models (and other industry models) focused on replicating “last-second, reactive” human maneuvers, according to the company.

Exclusive-Uber’s Commitment to Self-Driving Startup Nuro Is Close to $500 Million, Sources Say (link).

Waymo Readies First National Ads as Rivals and Critics Proliferate (link, no paywall). I liked the ad.

Autonomous vehicles were supposed to cut traffic—what if they don’t? (link). Not sure I agree with the title (I certainly never thought that) but the ‘paper’ the article is about seems to echo an article we wrote on TDD last year (link).

A similar analysis conducted late last year on Waymo’s CPUC data from January 2024 through September 2025 by Matthew Raifman, who studies policy and autonomous vehicles at UC Berkeley, also found that 44 percent of Waymo’s miles were driven with empty vehicles and that two-thirds of those empty miles were robotaxis driving around waiting to be assigned a customer... In total, about 40 percent of the miles traveled by a Lyft or Uber driver are deadhead miles, suggesting there’s little difference in congestion whether there’s a human behind the wheel or not.

Lockton and Nexar Introduce Human-Benchmark Framework for Autonomous Vehicle Safety (link).

‘When we talk about autonomy, everyone looks at the vehicle, but the real work is in the support layer. You can’t have a fleet without a foundation that’s ready to handle the day-to-day grind of thousands of units (link)’ via Nick Allen at TaskUs.

What else we’re reading/listening to

Autonocast by , , and Kirsten Korosec: #365: How To Insure Autonomous Vehicles w/ Steve Miller of Hub International (link).

Europe’s First Driverless Vehicle Approval with Bliq CEO Julian Glaab (link). Our latest podcast where we discussed Bliq’s transition from ride hailing to autonomous vehicles, why the company is betting on privately owned AVs, and what Europe’s evolving regulatory landscape means for the future of autonomy, among other things.

Autonomy Insiders by , with : What Waymo’s Data Reveals About Who Actually Rides Robotaxis (link).

AVs/Humans behaving badly

‘I schedule a Waymo for 4:15, it arrives at 4:09 and is leaving 2 mins before the scheduled arrival time? And then they charge me for being on time??’ (link). This is a poor pickup experience and an easy way to damage the brand. Scheduled rides are often reserved for high-stakes trips like airport runs and important meetings, which is why riders pay a premium and expect the vehicle to arrive early and to wait. Having a Waymo show up ahead of schedule, wait just two minutes, and leave before the actual pickup time is a big miss. If the issue was that it was stopped in a bike lane, the car could have circled the block and returned closer to the scheduled time. It feels like a relatively easy fix, and it’ll be interesting to see how quickly Waymo addresses it.

‘Waymo decides to drive on the wrong side of the road’ (link).

‘It is insanity’: SF luxury condo owners fume over aggravating neighbors - They’re Waymos (link, no paywall).

AVs/Human behaving “goodly” :)

‘A good example of predictive AI from the Waymo Driver: in a split second, it predicted an initial crash of human-driven cars in the next lane, anticipated a secondary collision, and proactively shifted lanes — all while managing its distance from the vehicles around it’ (link).

‘When riding in the Ojai for the first time it plays an updated welcome video showing how to manually open the doors in case of an emergency’ (link).

May Mobility: Nagoya’s streets don’t leave much room for hesitation. Tight streets are exactly where reasoning in real time earns its keep (link).

Shout-outs

Big thanks to TDD readers at Zag Daily for referring new subscribers. If there’s someone you think would enjoy TDD, feel free to forward this email or use the referral button below, and we’ll make sure to shout out your company.

Refer a friend

Neat Jobs

Fleet Operations and Field Engineering Lead at Wayve (link) via Ken M.

Vehicle Recovery, Safety, & Logistics - EAST at Waymo (link) via Luke Gunderson.

Job Moves

Jillian Noël: Piva Capital -> Terawatt (link).

Adam Warner: EverDriven -> TaskUs (link).

Leticia Cavalcante: Waymo (link).

We’re now at 11,000+ subscribers! Interested in advertising with TDD? Check out our media kit here and reach out to us. Newsletter spots are currently sold out until September.

Reach out

Until next week :)

-Harry