Intro to Physical AI
AI is leaving the screen — into robots, vehicles, and machines that act in the real world. A guided tour of Physical AI from a core contributor to NVIDIA's Cosmos World Foundation Model.
Led by Nalin Dadhich
What it's about
For a decade, AI lived on screens — text, images, predictions. The next wave moves into the physical world: robots that manipulate, vehicles that drive, machines that perceive their surroundings and act. The industry calls it Physical AI.
This session is a working map of that shift: what physical AI actually is, the stack underneath it — perception, reasoning, action, simulation, synthetic data, world models, robotics — and an honest read on what's in production versus what's still a demo reel.
Nalin builds at the center of it: he's a core contributor to Cosmos, NVIDIA's open world foundation model for physical AI. Expect a practitioner's view — what's genuinely working, the hard problems still open, and what the shift means for anyone building or investing in the space.
What you'll take away
- A clear definition of Physical AI — and why it's distinct from the AI you already use.
- The stack that makes it work: perception, simulation and synthetic data, world models, robotics.
- Where physical AI is real in production today — and where it's still demo-stage.
- The industries physical AI reshapes first — and how.
- The hard, unsolved problems — sim-to-real, data, physical plausibility, safety — that gate progress.
- What the shift means for founders, operators, and investors eyeing the space.
Who should attend
Founders & builders
Exploring robotics, autonomy, or embodied-AI opportunities.
Operators & investors
Sizing up where physical AI is real versus hype.
The technically curious
Senior folks who want a grounded map of the next AI frontier.
Who'll be leading
Nalin Dadhich
Senior Machine Learning Engineer, NVIDIA
Nalin Dadhich is a Senior Machine Learning Engineer at NVIDIA and a core contributor to NVIDIA's world foundation models for Physical AI — Cosmos.
His work spans LLMs, computer vision, model optimization, synthetic data generation and the perception layer that physical AI depends on.
Live on Zoom · 60 min
July 26 · 10:00 AM IST