Subvolts take Physical AI needs better data tools; Rerun's game tech offers a solution. Dec 16, 2025 - Developers
- Learn how Rerun's ECS data model tackles complex sensor data challenges for physical AI.
- Business
- Understand the critical role of specialized data tooling in enabling real-world AI applications.
This interview provides insight into specialized developer tools that are critical for advancing physical AI applications.
Why Physical AI Needed a Completely New Data Stack
Audio streams from Lukas Biewald. Your place is remembered on this device.
From Subvolts
Summary and analysis by Subvolts editorial, AI-assisted.
Lukas Biewald interviews Nikolaus West, CEO of Rerun, on the challenges of bringing AI into the physical world. West explains how Rerun uses an Entity Component System (ECS), a game development data model, to manage complex sensor data. This approach aims to simplify tasks like flexible manipulation, moving AI beyond the lab.
AI needs to understand the real world, which generates messy, constantly changing data from sensors. This episode explains how a company called Rerun uses a data system originally designed for video games to organize this complex information, making it easier for AI to work in physical tasks like robotics.
This episode is for AI developers and engineers working on robotics or physical AI applications. It highlights how specialized data tooling, like Rerun's ECS approach, is crucial for handling the messy, multimodal, and time-aware data from real-world sensors. This directly addresses a bottleneck that frontier labs like Google Gemini and OpenAI face as they push AI into physical domains, moving beyond purely digital tasks.
- HostLukas Biewald
- GuestNikolaus WestCEO, Rerun
- Guest: Nikolaus West, CEO of Rerun
- Show: Gradient Dissent
- Length: 61 minutes
- Data Model: Entity Component System (ECS)
- Application Area: Physical AI, flexible manipulation
Chapters from the publisher ↗ Key moments and transcript search come with a later piece.
The future of AI is physical.
In this episode, Lukas Biewald talks to Nikolaus West, CEO of Rerun, about why the breakthrough required to get AI out of the lab and into the messy real world is blocked by poor data tooling.
Nikolaus explains how Rerun solved this by adopting an Entity Component System (ECS), a data model built for games, to handle complex, multimodal, time-aware sensor data. This is the technology that makes solving previously impossible tasks, like flexible manipulation, suddenly feel "boring."
Connect with us here:
Nikolaus West: https://www.linkedin.com/in/nikolauswest/
Rerun: https://www.linkedin.com/company/rerun-io/
Lukas Biewald: https://www.linkedin.com/in/lbiewald/
Weights & Biases: https://www.linkedin.com/company/wandb/
- Entity Component System (ECS)
- Nikolaus West
- A way to organize data, originally used in games, that's good for handling lots of changing information from sensors.
- multimodal
- Nikolaus West
- Involving or using several different methods or types of data, like text, images, and sensor readings.
1 h 341 h 15He's Building an AI That Can't Lie | Dan Klein
WithDan Klein
1 h 15He Raised $70M to Cure Every Disease With AI | Samuel G. Rodriques, Edison Scientific
WithSamuel G. Rodriques
42 min
1 h 40AI 2025 → 2026 Live Show | Part 2
WithAlex Boris, Dean Ball, Peter Wildeford
1 h 55Sarah Paine — Why Russia Lost the Cold War
WithSarah Paine
1 h 55AI 2025 → 2026 Live Show | Part 1
WithZvi Moshowitz, Greg, Eugenia Kuyda, Ali Behrouz
Questions people ask
- What is the main challenge for physical AI?
- The breakthrough needed to get AI out of the lab and into the real world is blocked by poor data tooling for complex, multimodal, time-aware sensor data.
- How does Rerun address this challenge?
- Rerun adopted an Entity Component System (ECS), a data model built for games, to handle complex sensor data, making tasks like flexible manipulation easier.
Page generated Oct 19, 2026. Summaries are Subvolts' own; the audio, artwork and notes belong to Lukas Biewald.