The Full Stack of Physical AI — CVPR 2026 Tutorial

Co-organizer of “The Full Stack of Physical AI: Simulation, Foundation Models, and Edge Deployment for Next-Generation Robotics Applications” — a half-day tutorial at CVPR 2026 in Denver, Colorado.

Tutorial page: johnnynunez-nv.github.io/physical-ai-cvpr2026

Overview

This tutorial provides a hands-on, end-to-end walkthrough of the full Physical AI stack — from data collection and simulation to foundation model training and real-time edge deployment. Attendees build a complete robotics pipeline using NVIDIA’s state-of-the-art tools and hardware.

Key Topics

• Simulation & Synthetic Data — Isaac Sim, Isaac Lab, and Newton physics engine for scalable robot learning environments
• Foundation Models — GR00T, ACT, and Vision-Language-Action (VLA) models for robot manipulation and locomotion
• Edge Deployment — Hardware-aware inference optimization on NVIDIA Jetson Thor for real-time robot control
• Human-in-the-Loop — Teleoperation, data annotation, and fine-tuning workflows

What Attendees Take Home

• Slide decks and training datasets
• Full pipeline code and open-source repositories
• Deployment guides for each component
• Community Discord access

Organizers

• Dr. Raymond Lo* — Developer Advocate Manager, NVIDIA
• Johnny Núñez* — Developer Advocate, NVIDIA
• Chitoku Yato — Sr. Technical Product Marketing Manager, NVIDIA
• Spencer Huang — Product Lead for Robotics, NVIDIA
• Dr. Mitesh Patel* — Sr. Developer Advocate Manager, NVIDIA

* Equal contribution