Physical AI: Teaching Machines to Understand the Real World

62 snips

Feb 6, 2026

Nick Gillian, Co-Founder and CTO of Archetype AI and builder of Newton, is an expert in real-time sensor understanding. He discusses Physical AI: single foundation models for diverse sensors, fusing non-visual modalities, and the engineering challenges of massive continuous data. He also covers evaluation strategies, product-driven dataset design, and turning research into deployable, safety-minded systems.

Ask episode

AI Snips

Chapters

Transcript

Episode notes

INSIGHT

Physical AI Is Much Broader Than Robots

Physical AI covers any use case where sensors map the real world into reasoning and action across industries.
Archetype aims to build a horizontal foundation model that customers plug diverse sensors into for domain-specific applications.

INSIGHT

One Mothership Model, Many Deployed Slices

Archetype builds one large foundation model that supports many sensor types and is steered by natural language.
Customers typically slice and compress the base model to deploy a smaller, task-specific runtime in constrained environments.

INSIGHT

Sensor Models Need New Architectures

Techniques for language and vision models don't directly transfer to arbitrary sensor modalities like LiDAR or hundreds of time series.
Newton requires new architectures, synchronization, and dataset strategies to ingest non-internet sensor data.

Get the Snipd Podcast app to discover more snips from this episode

Get the app

Nick Gillian is the Co-Founder and CTO at Archetype AI, working on physical AI foundation models that understand and reason over real-world sensor data.

Physical AI: Teaching Machines to Understand the Real World // MLOps Podcast #360 with Nick Gillian, Co-Founder and CTO of Archetype AI

Join the Community:

https://go.mlops.community/YTJoinIn

Get the newsletter: https://go.mlops.community/YTNewsletter

MLOps GPU Guide: https://go.mlops.community/gpuguide

/ Abstract

As AI moves beyond the cloud and simulation, the next frontier is Physical AI: systems that can perceive, understand, and act within real-world environments in real time. In this conversation, Nick Gillian, Co-Founder and CTO of Archetype AI, explores what it actually takes to turn raw sensor and video data into reliable, deployable intelligence.

Drawing on his experience building Google’s Soli and Jacquard and now leading development of Newton, a foundational model for Physical AI, Nick discusses how real-time physical understanding changes what’s possible across safety monitoring, infrastructure, and human–machine interaction. He’ll share lessons learned translating advanced research into products that operate safely in dynamic environments, and why many organizations underestimate the challenges and opportunities of AI in the physical world.

// Bio

Nick Gillian, Ph.D., is Co-Founder and CTO of Archetype AI with over 15 years of experience turning advanced AI and interaction research into real-world products. At Archetype, he leads the AI and engineering teams behind Newton—a first-of-its-kind Physical AI foundational model that can perceive, understand, and reason about the physical world. Before co-founding Archetype, Nick was a Senior Staff Machine Learning Engineer at Google and a researcher at MIT, where he developed AI and ML methods for real-time sensor understanding. At Google’s Advanced Technology and Projects group, he led machine learning research that powered breakthrough products like Soli radar and Jacquard, and helped advance sensing algorithms across Pixel, Nest, and wearable devices.

// Related Links

Website: https://www.archetypeai.io/https://www.archetypeai.io/blog/timefusion-newton https://www.nature.com/articles/s41598-023-44714-2 https://www.youtube.com/watch?v=Pow4utY9teU https://www.youtube.com/watch?v=uE0jjdzwe9w https://arxiv.org/abs/2410.14724

Coding Agents Conference: https://luma.com/codingagents

~~~~~~~~ ✌️Connect With Us ✌️ ~~~~~~~

Catch all episodes, blogs, newsletters, and more: https://go.mlops.community/TYExplore

Join our Slack community [https://go.mlops.community/slack]

Follow us on X/Twitter [@mlopscommunity](https://x.com/mlopscommunity) or [LinkedIn](https://go.mlops.community/linkedin)]

MLOps Swag/Merch: [https://shop.mlops.community/]

Connect with Demetrios on LinkedIn: /dpbrinkm

Connect with Nick on LinkedIn: /nick-gillian-b27b1094/

Timestamps:[00:00] Physical Agent Framework[00:56] Physical AI Clarification[06:53] Building a Repair Model[12:41] World Models and LLMs[17:17] Data Weighting Strategies[24:19] Data Diversity vs Quantity[38:30] R&D and Product Creation[41:22] Construction Site Data Shipping[50:33] Wrap up