Autonomy Expansion: Rivian Buys 15-Person UK Firm for 3D Road Tests
The company absorbed a small London team to process over 400,000 daily video clips into virtual environments for its self-driving software.
By Alina Carter ·
Rivian has acquired Third Dimension, a 15-person startup based in the United Kingdom, to convert real-world driving footage into 3D simulation environments. The deal, first reported by TechCrunch, remains unpriced as the automaker did not disclose the purchase amount. The acquisition is part of a strategy to test its autonomous driving systems against rare and complex road scenarios without waiting for them to occur naturally on the street.
The technology relies on data from Rivian's vehicle fleet. Owners of the R1S, R1T, R2, and commercial vans who opt into camera sharing generate more than 400,000 driving sequences daily. While most of this footage shows standard commuting, autonomous systems struggle with edge cases. These include a mattress in a fast lane, a police officer directing traffic through a red light, or a child running between parked cars. The new team will digitize these rare events to create controllable test beds for engineers.
Third Dimension was founded in 2024 by Tolga Kart and Piotr Sokólski. Kart spent more than two years working on Autopilot simulation at Tesla and later led engineering at Parallel Domain. Sokólski spent four years at Wayve building photorealistic neural simulators. The startup raised a $7 million seed round led by Felicis in October 2024. It originally targeted customers in gaming, film, and the US military, but its new focus is squarely on autonomous vehicles.
All 15 employees from Third Dimension are moving into Rivian's autonomy and AI group, which is led by James Philbin. The hire expands the company's operational footprint in London. Philbin stated that simulation is central to training and validating the system. He described the acquired technology as a tool that turns real-world data into detailed 3D environments, allowing engineers to alter variables in specific scenes rather than waiting months for natural occurrences.
The move places Rivian in a crowded field of automotive simulation. Waymo unveiled its World Model in February, built on Google DeepMind's Genie 3, to generate unencountered scenarios. Tesla has also demonstrated a neural world simulator for its Full Self-Driving software. These tools aim to solve the problem of rare events by generating synthetic data that mimics real conditions.
Financial stakes drive the urgency behind the engineering push. In March, Uber agreed to invest up to $1.25 billion in Rivian to deploy R2 robotaxis. Only $300 million is paid upfront. The remaining funds are tied to autonomy milestones, including the deployment of 10,000 driverless R2 vehicles in San Francisco and Miami starting in 2028. This timeline pressures the autonomy team to accelerate development and testing cycles.
Consumer timelines for autonomy features have shifted recently. The point-to-point driving feature is scheduled for a small group of invited Autonomy+ owners later this year, under a nondisclosure agreement. The "eyes-off" driving capability, which CEO RJ Scaringe promised for 2026 in January 2025, is now targeted for 2027. Level 4 capability is planned for 2028 on the Gen 3 hardware shown at the company's AI and Autonomy Day event.
