Since I previously shared how we are scaling AI usage so quickly without costs rising at the same rate, there’s been a lot of interest in the engineering details. So I asked the team to open up the hood and share what we’ve learned. One thing we’ve always been very good at @Uber is understanding the economics of a system at a deep level. And I’ve always believed that constraints drive innovation. We took the opportunity to turn the cost problem into an engineering problem. We broke AI spend down into its underlying levers and started systematically attacking each one: • Scaling vendor-neutral managed agents • Building Uber SWE benchmarks to measure what actually works • Optimizing prompt caching • Making tool and MCP usage more efficient • Grounding trajectories in our context graph • Building efficient, reusable agent skills • Giving engineers real-time cost visibility and optimization tips There’s a lot more work to do in this space: scaling more autonomous agents, dynamic model routing, a self-evolving context graph, and continuously updating efficient agent skills. There's still a tremendous amount to invent here, and we’ll keep sharing what we learn as we build and deploy it at Uber. This is an incredible time to be an engineer. The economics of software are changing, the way we build is changing, and engineers have an opportunity to help define what comes next. Read the deep dive here: and let us know what you think.

It was great to visit our @Uber AV Labs team with @sachinkansal and see firsthand the progress we’re making to accelerate the entire AV ecosystem. We launched AV Labs earlier this year with a simple goal: to build the world’s most geographically diverse, high-quality dataset for autonomous driving and make it available to our growing network of 30+ AV partners. Working with @ROUSHPerf, we’re equipping a fleet of vehicles with 14 cameras, 8 solid-state lidars, and 9 radars. These sensors capture a 360 degree, time-synchronized view that can help our AV partners train and improve their self-driving tech across a wide range of real-world environments and conditions. This year, we plan to deploy up to 500 of our data collect vehicles globally, with the first hitting the road this summer. These vehicles will collect data while completing real trips on the Uber platform with Uber drivers, helping us capture the complexity of real-world driving at scale. By combining the scale of our platform with the expertise of our partners, we can accelerate progress across the entire AV ecosystem. Proud of the team making this vision a reality! @UberEng

As our CFO @_balaji_km mentioned at earnings today, we’re seeing some very interesting trends on AI costs. I think it’s another signal that we’re coming to the end of the so-called ‘tokenmaxxing’ era. Here’s what’s been happening behind the scenes. Since the beginning of the year we’ve more than quadrupled the number of people using frontier AI tools. That’s thousands of engineers using them every single day. During that same period, our cost per token has declined. You might expect costs to rise as adoption accelerates. We've seen the opposite. Not because we've restricted access, but because we've treated efficiency as an engineering problem rather than a budget problem. A few examples: • Caching and reuse: We use optimizations to improve our prompt cache hit rate that reduce our input token spend. • Better defaults and tooling: We tuned default model settings, context sizes and developer workflows so teams get the same results with fewer tokens and lower-cost inference. • Visibility drives efficiency: We gave engineers real-time visibility into their AI usage and costs per hour. • Experimenting with open-weight models: we continuously evaluate new models and deploy the best option for each use case. This is the future of applied AI at enterprise scale. The next phase, whatever we call it, will not be characterized by who spends the most tokens, but about how people use them as efficiently as possible. Credit to all the engineers at @Uber who are helping to build this future. 🚀

Last week, we celebrated our interns in honor of #NationalInternDay# with a dedicated day of events, and it was a great reminder of the energy and talent that make our program at @Uber so special. I had the opportunity to join a panel alongside three former interns who are now software engineers on my team, and saw firsthand the immense creativity at this year's intern hackathon. Spending some time today with this group left me more convinced than ever that the future of engineering is bright! @UberEng

AI agents are everywhere at @Uber. It’s great to see, but the thing that keeps me up at night is how we are going to secure them. This is something that I have been thinking about for a while. Today, our agents run 50,000+ sessions per day across thousands of endpoints. And this isn't just engineering anymore. Employees across the company use agents that read code, run commands, call internal tools, analyze data, and act on real systems. That scale forced us to confront an important question: How do you secure agents when your security tools can't even see them? Traditional Endpoint Detection & Response (EDR) sees the file write, but not the prompt that triggered it. It sees the network call, but not the agent's reasoning. The intent, the thing that separates malicious from benign, is invisible. So we built Agentic Detection and Response (ADR): • Capture the full causal chain: prompt → reasoning → tool call → outcome, across Cursor, Claude Code, Codex, and every agent our employees use. • Triage cheaply: a fast, high-recall first pass handles the flood of benign sessions. • Reason deeply: only suspicious events get expensive LLM analysis, enriched with source code, threat intel, and policy context. • Red-team continuously: an offline explorer evolves hard attack variants before attackers find them. After 10+ months in production, the results speak for themselves: • Hundreds of credential exposures detected across 26 categories. • Shift-left prevention blocking secrets at 97.2% precision, before they ever leave the laptop. • Zero false positives on our enterprise benchmark, with 2-4x the F1 score of state-of-the-art baselines. • Every attack detected on AgentDojo, the public prompt injection benchmark. Just as valuable as the detections are the lessons from running this in production: • The workflow is the unit of security, not the individual tool call. Attacks hide in causally-linked chains that look benign step by step. • Credential leakage is a far more common operational issue than prompt injection. • Approval fatigue is real: when users approve 50+ actions per session, human oversight becomes a rubber stamp. You can't secure agents you can't observe. And nobody can solve this alone. That is why we recently joined the Open Secure AI Alliance (OSA), and why today we're taking the next step: open-sourcing ADR. The release includes the ADR Sensor, the detection framework, and ADR-Bench, the first enterprise agentic AI security benchmark: 302 tasks derived from real production telemetry and full coverage of all 17 attack techniques across 5 tactics, so the community can rigorously evaluate their own defenses. Code: Paper: The future of AI security won't be built behind closed doors. Excited to see what the community builds on it, and what we all learn together! @UberEng

Great to see more than 230 organizations across the AI ecosystem have signed a letter supporting open weights as part of America’s AI future.  Their message: open weights enable more people to build, compete, and put AI to work.  Here’s why that matters. 🧵

As AI becomes core infrastructure, security can't be an afterthought, and it has to be built in from day one. Excited for @Uber to join the Open Secure AI Alliance and collaborate with the industry to help build secure, trustworthy, and open AI for everyone.

Earlier this week, our @UberEng teams came together for Dispatch—an incredible day of collaboration, learning, and connecting across our engineering sites around the world. We had an outstanding fireside chat with @bcherny, creator of Claude Code, alongside Uber's @hudaman. Their conversation explored how AI is fundamentally changing software development—not by replacing engineers, but by enabling them to solve harder problems, move faster, and build in entirely new ways. Having grown up at @Uber, one lesson has shaped how I think about building: the biggest impact comes from obsessing over hard, meaningful problems that matter to the business. Some of our best engineers spent years solving problems at Uber scale, open-sourced the technologies they created, and changed the industry in the process. Many later went on to build category-defining companies. It was a privilege to welcome back @SamarAtTemporal (Co-Founder & CEO of @temporalio), @martin_c_mao, (Co-Founder & CEO of @chronosphereio), and @byte_array (Founder & CEO of @Onehousehq). Their journeys all reinforce the same idea: don't start with "I want to build a company." Start with "I want to solve an important problem." Stay obsessed with it, keep building, and the rest can follow. Huge thanks to everyone who presented, participated, and helped make Dispatch such an inspiring event. The energy, curiosity, and technical depth across our teams make me incredibly excited for what we'll build next.

At @Uber, we support a strong American AI ecosystem across both frontier and open-weight models. We see them as complementary: the opportunity is to use the right model for each task, balancing capability, customization, cost, and control.

An exciting milestone for @Uber this week as we announced an offer to acquire @deliveryherocom. I’ve had the opportunity to spend time with Delivery Hero’s tech leadership team, and I’ve come away very impressed. They’ve built an exceptional engineering team and products that millions of people use every day. Together, we have a fantastic opportunity to create tech that scales globally while continuing to flourish locally. There’s a lot of work ahead, and the two companies will very much continue to compete until the transaction is finalized. But I look forward to welcoming the team and getting to work!

Agentic AI adoption is on fire at @Uber, and it's changing the way we build, not just in engineering, but across the entire company. Today, 99% of our engineers use AI tools. More than 70% of pull requests are attributed to local or cloud agents. And our engineers have built 2,500+ agent skills across the software development lifecycle. Those numbers are exciting, but they led us to a much bigger question: How do we bring agentic AI beyond engineering? Finance. Legal. Operations. Marketing. Customer Support. HR. Procurement. These functions run on complex workflows that are often manual, highly nuanced, and spread across dozens of systems. You can't automate them effectively by looking at process diagrams or documentation. You have to understand how the work actually gets done. So we created something called Agentic Pods. The idea is simple. We handpicked ~30 of our most AI-proficient engineers (people with deep knowledge of Uber's systems) and paired each of them with a domain expert from a business function. Then we gave every pod just two weeks. • Days 1 – 2: Shadow the expert. Observe every step. Document workflows. Ask questions. Build intuition. • Day 3: Prioritize opportunities based on scale, repetition, business impact, and data availability. • Days 4 – 5: Build a working agent alongside the person doing the job. • Days 6 – 9: Validate with several others performing the same work. Does it generalize? Does it actually make their job better? • Day 10: Ship. In just the past two months, we've run 16 Agentic Pods across 16 different business functions. • Capital allocation across 150 cities: 15 hours → 30 minutes. • Financial pacing reports: 2 days → 10 minutes. • Marketing web quality assurance: 2 weeks → 50 minutes. • Support workflow creation: 9,000 manual workflows → self-service automation. The productivity gains are impressive, but what surprised us most wasn't the speed. • It was how quickly engineers embedded in unfamiliar domains uncovered opportunities that had been hiding in plain sight. • The biggest wins rarely come from automating one task. They come from rethinking an entire workflow. Once you redesign the workflow around AI, you often eliminate handoffs, remove unnecessary approvals, replace legacy tooling, reduce vendor spend, and dramatically accelerate decision-making. • The workflow becomes the unit of automation - not the individual task. • The most impactful agent skills cut across teams, orgs, functions, tools, and systems. The biggest lesson? The best AI opportunities are rarely visible from the outside. You discover them by sitting next to the people doing the work, understanding every friction point, and building with them, not for them. We're now forming a dedicated team to scale this further and go deeper. They'll deeply understand the work, redesign it from the ground up, and use AI to fundamentally change how the business operates. It's exciting times!

Excited to welcome three exceptional leaders to Uber's tech team: Mattie Toia (VP of Engineering, Infrastructure), @JohannesGehrke (AI Fellow), and @SecurityGuyPhil (Chief Information Security Officer). As we invest in AI, infrastructure, and security, their leadership will help us build the next generation of technology that powers millions of trips, deliveries, and businesses every day. The future of Uber will be defined by technical innovation and relentless execution. We're just getting started. 🚀