Adi Bathla, CEO and founder of Revv – Interview Series

Adi Bathla, CEO and founder of Revv, is a product-driven operator and innovation leader based in New York who has built his career at the intersection of technology, systems thinking, and scale. Prior to founding Revv, he led product and digital customer experience initiatives at high-growth commerce companies, helped launch new business lines within large enterprise platforms, conducted research on artificial intelligence and collective intelligence at MIT Sloan, and earlier in his career led award-winning space systems design teams connected to NASA research programs.

Revv is an AI-powered automotive repair platform focused on simplifying ADAS calibration and diagnostics for collision and mechanical repair shops. By combining OEM-grade documentation, intelligent workflows, and deep integrations with existing shop and estimating systems, Revv helps repair centers reduce manual research, improve compliance and safety, and turn increasingly complex calibration requirements into scalable, data-driven operations. As vehicles become more software-defined, Revv is positioning itself as core infrastructure for modern repair workflows across North America.

Your early exposure to the auto repair industry clearly shaped your path. Can you share a specific moment from that time that made you realize this space needed an AI-driven solution?

There was this one phone call that completely changed my view of the industry. A shop owner called me in a panic: he’d repaired a car, but the lane change system malfunctioned afterward, and he was terrified of getting sued. That moment made me take a deeper look into ADAS systems, and I realized this invisible complexity was a massive problem shops couldn’t solve on their own. Since ADAS repairs weren’t obvious like a dent or scratch it was easy for them to go unnoticed. Technicians were spending 3-4 hours just documenting work and finding repair procedures, like they were searching for a needle in a haystack. That’s when I knew AI could cut through all that noise and give technicians exactly what they need in seconds instead of hours.

Your time working on AI and collective intelligence research at MIT and earlier systems-level work at NASA exposed you to complex, safety-critical environments early on. How did those experiences directly influence your decision to found Revv and focus on vehicle calibration as a software problem?

My experiences at NASA and MIT taught me that to build and be a successful entrepreneur, you just need to learn fast and surround yourself with the right experts. This mentality gave me the confidence to walk into an industry I knew very little about and challenge the way things had traditionally been done.

When I started spending time in auto shops, I saw technicians drowning in manuals, trying to calibrate sensors that control whether a car can brake properly or stay in its lane. It reminded me of the safety-critical environments I’d been exposed to at NASA, where precision is key. This is life saving technology, but shops were managing it with paper procedures and outdated systems. I quickly realized this was a software problem disguised as a mechanical one. Cars had become computers on wheels, but the infrastructure to service them hadn’t caught up. That brought me back to the entrepreneurial mindset I developed at MIT and NASA, and it’s how we built Revv: pairing an outsider’s fresh perspective with deep expertise from the technicians actually doing the work.

Before Revv, you led product and innovation initiatives at Jet and Walmart at massive scale. What lessons from building 0→1 products in commerce carried over most clearly when designing software for the automotive repair ecosystem?

The biggest lesson I learned from my time at Walmart was about meeting users where they are, not where you want them to be. I was building for suppliers and manufacturers who weren’t tech-savvy and had been doing things the same way for decades. You can’t ask them to replace their entire system. Instead, you embed your expertise into their existing workflows so they don’t have to lift a finger outside of what they’re already doing.

That became Revv’s entire thesis. We integrate with shops’ existing tools and software, run in the background, and deliver insights without disrupting what’s already muscle memory for them. But I also carried over what I learned at Jet about talent: A+ players bring A+ players, and you need to be obsessive about building the right team from day one. I ingrained this as our hiring practice at Revv, because none of the work we do would be possible without the team of allstars we’ve built.

Automotive repair is one of the largest yet least modernized industries in the U.S. When you first started building Revv, what resistance or skepticism did you encounter, and how did you overcome it?

The resistance at first was a huge challenge because shops have been doing things the same way for over 40 years, and change is uncomfortable. But here’s what worked: I didn’t just pitch them software, I committed to understanding every single pain point. I handed out cards with my phone number and said, “if you have a problem, call me.” And they did. I spent years building that database of trust and knowledge.

The breakthrough was showing them we weren’t asking them to replace their systems or change how they work. We built Revv to integrate directly into their existing software and workflows, running in the background and delivering what they need without disrupting their current processes. Once shops saw we actually understood their world and were making their jobs easier, not harder, the skepticism started to fade.

Revv positions itself as an operating system for software-defined vehicles rather than a single point solution. What does being an operating system mean in practical terms for calibration shops and repair networks?

It means we’re not just solving one problem, we’re becoming the infrastructure that powers their entire ADAS workflow. A car comes into the bay, Revv connects to their existing tools, pulls data directly from the manufacturer, and delivers a complete package to the technician in seconds. It gives them step-by-step repair instructions, every required calibration, original equipment manufacturer documentation, and the claims packet ready to submit to insurance.

We’re evolving from a system of record to one of action, not only telling them what needs to be done, but handling the administrative work for them. By 2025, over 74% of our users are adopting our new products because they see us as the single platform that handles their entire ADAS operation end-to-end. This tracks with what we’re seeing industry-wide. Our recent ADAS Benchmark Survey of 300 autobody professionals found that in-house calibrations are expected to grow from 57% to 64% over the next two years.

Cars are now rolling computers packed with sensors, cameras, and software dependencies. Where do human technicians struggle most today, and how does AI meaningfully support their decision-making without taking control away from them?

Technicians struggle most with the administrative burden that now comes with modern repairs. From calibrating sensors to tracking down manufacturer manuals and putting together insurer-ready reports, every step has to be researched, documented, and approved, turning a hands-on job into one filled with paperwork. Every estimate has 100-200 line items, and each one has a ripple effect. Back in 2023, the average repair required two to three calibrations. Now it’s over five. Technicians are spending three to four hours just documenting work and hunting down procedures, but with Revv and AI, that process drops to three to five minutes.

Revv’s AI processes all of that complexity in the background, connecting directly to manufacturer data, identifying every required calibration, and delivering step-by-step instructions. We’ve processed over 300,000 repairs, with over 5,000 customers now using the platform. For documentation, technicians can submit photos and Revv generates the entire insurance-ready report automatically. Our platform flags potential missed steps and automates the repetitive work, but the technician stays in full control making the decision without the administrative burden.

Revv’s models are trained on hundreds of thousands of real repair events. How do you ensure data quality, accuracy, and compliance when AI recommendations are tied directly to safety-critical outcomes?

For us, data quality and accuracy start with grounding the AI in real-world repair experience, with our models built directly on insights from experienced technicians across multiple regions and vehicle types.

We also build in continuous feedback loops, so technicians can validate AI recommendations in real time. Every calibration and procedure is cross-checked against the exact OEM manuals and technical documentation for a specific vehicle. With a database of over 300,000 repairs from two countries, our platform keeps learning and improving, while technicians stay in control throughout the entire process.

Revv works across calibrators, repair networks, insurers, and OEM systems. How do you design an AI platform that creates trust and value across stakeholders with very different incentives?

We think of Revv as the connective tissue between technicians, insurers, and consumers, so we work to meet all of their collective needs.

For technicians, we’re saving hours of admin time while helping them capture missed revenue by identifying calibrations that would’ve slipped through the cracks. Insurers get faster approvals, accurate documentation, and fewer disputes. Consumers get their cars back safer and faster because we’re ensuring every required calibration is actually completed.

As vehicles continue to evolve into fully software-defined platforms, what does success look like for Revv three years from now, and what capabilities will repair infrastructure need to develop to keep pace?

To keep pace, shops will need in-house capabilities, a pipeline of next-generation technical talent, and a strong collaboration network to ensure that every repair is accurate and efficient.  By 2029, regulations will require all new car models to have emergency braking, and shops are starting to see how much value doing in-house calibrations brings to their businesses. In our recent survey, 74% of autobody professionals now report ADAS as a profit generator, with 60% considering growing ADAS revenue ‘extremely or very important.’

What we’re already seeing is ADAS calibration becoming its own category, with new specialists emerging every month and real business momentum building around it. As we look ahead, we see Revv serving as the backbone of this entire ecosystem. That means the platform becomes the standard across collision shops, giving technicians, insurers, and customers one unified system to manage and deliver safe, compliant calibrations at scale. We’re building the infrastructure that defines how software-driven vehicles get repaired, and ultimately, we’re setting the standards that will shape the industry’s future.

For founders bringing AI into deeply entrenched, traditionally analog industries, what common mistakes do you see, and what assumptions did you personally have to unlearn while scaling Revv?

Something I learned early on is to cut through the noise and focus on the problem first, not the solution. It’s easy to get caught up in the buzz and start building something impressive, but that often turns into a solution in search of a problem.

What matters is finding the issue that customers actually experience every day. The assumption I had to unlearn was thinking better technology alone would win. I underestimated just how entrenched workflows are in this industry. When we started Revv in 2022, I spent time in shops with technicians to see their workflows up close and understand what was holding them back. This taught me that real change comes not from implementing flashy tech or convincing shops to adopt a new way of doing things. It comes from embedding your solution so seamlessly into their existing workflow that they don’t have to change anything. You don’t ask them to change, you make their current way better.