Founder, CEO at @PhysicalAI. Tech leader and executive, interaction designer, scientist. Google, Disney, Sony before. 2019 National Design Award. TED speaker.
San Francisco & Bay Area, USA
Joined August 2011
- Tweets3.2K
- Following332
- Followers5.4K
- Likes2.2K
Ivan Poupyrev, PhD retweeted
Newton’s embeddings turn a window of raw sensor readings into a single point, placed so that moments that look alike in the physical world land near each other.
This public dataset, PAMAP2, shows data recorded when a group of nine people wore three motion sensors, on the wrist, the chest and the ankle, while going through twelve everyday activities.
This visualization tool helps our team see the data in more detail and isolate different states: Rest, walk, run, cycle, stairs and housework each settle into their own region.
Seeing why a model groups two moments together is the first step toward understanding what it tells you about the physical world.
Built with Newton on the PAMAP2 physical activity dataset (Reiss and Stricker, 2012).
Ivan Poupyrev, PhD retweeted
When people hear Physical AI, they picture a humanoid robot. Our CTO, Nick Gillian, makes the case that most of the real opportunity lies in existing equipment rather than in deploying a humanoid workforce.
Every factory is streaming radar, video, vibration, and current. Newton is one model that reads all of it, catches what’s off in real time, and adapts to machines it’s never seen, right at the edge.
Full conversation with @AI_in_the_AM ↓
Ivan Poupyrev, PhD retweeted
Our Chief Scientist @_jaimelien_ joins the panel at Long Horizon in San Francisco on Oct. 20 to take on one question: “Will World Models Eat Physical AI?”
“World model” is a broad term; some are built to generate or simulate the physical world. Newton is a physical world model that learns from real-world sensor data, so it’s built to understand what is actually happening on a machine or a factory floor.
That’s the view Jaime brings to the debate.
➡️ In SF? Join us: archetypeai.io/resources/eve…
Ivan Poupyrev, PhD retweeted
This advice is great for all Founders! I received it from my daughter on Father’s Day.
Ivan Poupyrev, PhD retweeted
Across 63 wind turbines, Newton found nine failure modes no one had seen before.
Anomaly Discovery Agent, powered by our world model, Newton, reads patterns across sensor streams and surfaces the issues no single signal stream can identify. Newton did this out of the box: no failure history to label first, and no model built turbine by turbine.
archetypeai.io/agents
Ivan Poupyrev, PhD retweeted
"Large language models may know a lot, but because they have been trained on text and images, that limits their understanding of our world. By contrast, Archetype AI’s Newton foundation model can interpret data from cameras, lidar, radar, inertial measurement units, and other sensors used in enterprise domains such as construction, logistics, manufacturing, and urban planning."
- @harrymccracken, @FastCompany
fastcompany.com/91612044/fou…
Ivan Poupyrev, PhD retweeted
Fantastic energy and engagement at AI4Energy Americas this week. Great discussions about how Physical AI can transform sensor data into a structured understanding of how physical assets behave—an essential step toward building the trust needed for closed-loop autonomy.
Thanks to SPE and @conocophillips for hosting. I’m looking forward to more conversations about how Physical AI is shaping the future of the energy industry.
Ivan Poupyrev, PhD retweeted
🗣️ Last call to register for our predictive maintenance webinar on Sept. 24.
🔗 luma.com/jwedb311
Ivan Poupyrev, PhD retweeted
BTS of AI Infra Summit as a speaker 👀 (what speakers talked about off stage)
Most companies don't have an AI problem. They have a context problem.
That was the loudest signal at AI Infra Summit this week. 3 takeaways from the rooms I was in:
Context is the real bottleneck.
AI can't do useful work without your company's internal knowledge, decision history, and workflows. The model isn't what's holding you back. Your data plumbing is.
Cybersecurity is the immediate risk.
AI makes attacks faster and cheaper. If your defenses are still manual, you've already lost the speed game. Defense has to be automated too.
Stop using one model for everything.
Frontier models for the hard problems. Cheaper models for routine work. The teams winning with AI treat model choice like a cost structure, not a loyalty.
Behind the scenes:
Great conversations in the speaker lounge with @theICToptimist (CTO, @DellTech), Chelle Izzi (Chief Commercial Officer, Pacific Gas and Electric Company), and @ipoupyrev, PhD (Founder, Archetype AI) among others.
Recorded a cross-pod with @LukasGentele, founder of @vcluster, and ran into @dylan522p (Founder, SemiAnalysis), who was also on Lukas's pod.
Good to catch up with friends and past podcast guests a Ray W. and @GopinathRashmi.
Also joined the AI Infra Summit live video as a guest.
Many thanks to the AI Infra Summit team for having me. More collabs coming soon.
Which of these three is your company underestimating right now?
#AI #AIInfrastructure #Cybersecurity
Ivan Poupyrev, PhD retweeted
I’ll be speaking alongside @ethanjb, Partner at @Venrock, at the AI Infra Summit on September 15. Ethan and I will have a candid conversation on what it takes to bridge digital intelligence and the real world, and what comes next.
We'll examine adaptive world models, models that learn directly from sensors, autonomous agents, and the growing impact of Archetype's Newton physical world model across industries.
See you there.
Ivan Poupyrev, PhD retweeted
The Manufacturing Leadership Council recently published a survey that validates some of my observations. 53% of organizations report that their data comes from incompatible systems and formats. Another 28% lack the skills to analyze their data effectively.
Counterintuitively, more data often means less clarity, and more sensors make it harder to make decisions.
Real-world industries need tools for turning data and observations from sensors into actionable understanding. That's the core vision behind our physical world model, Newton. It is pre-trained on billions of cross-modal sensor measurements (vibration, pressure, temperature, current, acoustics, radar, video, time-series telemetry, and more). It generalizes across machines, sites, and use cases without a custom model built for each one and delivers insights, predictions, and automation to our customers.
Newton eliminates those pain points and turns data into insights and decisions.
Full report:
manufacturingleadershipcounc…
Ivan Poupyrev, PhD retweeted
Newton Agent with skills are packaged instructions that teach a coding agent how to run a Newton task. Every agent blueprint also has a playground for testing against sample data.
➡️ Python SDK, agent skills, and the playground are all live: docs.archetypeai.app/for-age…
Ivan Poupyrev, PhD retweeted
We’re excited to have Naji on board at Archetype AI!
Naji brings operational infrastructure expertise from Scale AI, Vercel, and Google and joins us as our Sr. Executive Business Partner.
Welcome to the team 👏
Ivan Poupyrev, PhD retweeted
Great breakdown from @_jaimelien_ .
The standard approach to understanding how a machine works is to build a perfect digital replica in software, then use the simulation to learn its behavior.
But this is expensive, labor-intensive, and for complex systems, often impossible because the exact information to build that replica doesn't exist. And even if it did, your machine operates differently from the generic version. 🧵
Ivan Poupyrev, PhD retweeted
I just returned from a vacation that included a stop in Istanbul, a city layered with millennia of history. The artifacts I saw—some more than 2,000 years old—struck me because they reminded me that no two physical objects are ever exactly the same. Time leaves its own imprint on each one.
That is the essential difference between the physical and digital worlds.
Many algorithms assume that a model trained on one motor, press, or robotic arm should work on the same machine in another factory. Replicable. Deterministic. Software-like.
Reality is different.
No two motors spin identically. Every press behaves differently. Materials wear and shed atoms. Sensors drift and their signals distort. Each asset is shaped by its environment, maintenance, usage, and history. The physical world is always changing.
Trying to anticipate and account for every possible variation by collecting data in advance—even for a simple system—is prohibitively expensive and ultimately impossible.
With Newton, we are exploring a different approach: models that understand the fundamental physical behavior of systems, observe each machine, and then dynamically adapt to its particular behavior and conditions.
We call these adaptive local world models.
This is how we believe Physical AI can scale.
Ivan Poupyrev, PhD retweeted
📢 15 seats spots left for the first Newton Meetup on September 15. You will use your own data to build Physical Agents powered by our world model.
- Two-hour virtual session with hands-on support from Archetype AI
- Bring your own sensor or video data and a physical-world use case you actually care about.
- Built for ML engineers, data engineers, and technical leads
Afterward, you’ll have access to Newton for several weeks, plus a dedicated Slack channel with our engineers.
Come test a physical world model instead of reading about one: archetypeai.io/newton-meetup…
Ivan Poupyrev, PhD retweeted
Calling technical ML/AI and data engineers! 🗣️
We’re looking for 15 people who want to test and deploy our physical world model, Newton.
It’s BYOD (bring your own data), so you’ll use your own sensor and video data to build with Newton Agents and our API in real time.
Plus, after the meetup, you'll have continued platform access for several weeks and a dedicated Slack channel with access to the Archetype AI team to ask questions and get answers.
Think of this as a real opportunity to understand what Physical AI is capable of and apply it to your everyday job.
Save your spot here ⬇️
lnkd.in/gyEm-Tfp
Ivan Poupyrev, PhD retweeted
Congrats on joining the Archetype team, Komo and Dylan! 👏
Komo spent two decades at SANYO, MODEC, and SHIMANO shipping hardware and now brings Physical AI back to those makers as Head of Sales for Japan.
Dylan is an Earth scientist and remote sensing specialist joining Solutions Engineering.
Welcome to the team!
