The Pipeline
Robotics → Physical AI → NVIDIA Stack → Simulation → Intelligent Robots.
- 01Robotics
Sensing, control and autonomy
- 02Physical AI
Intelligence that has to act
- 03NVIDIA Stack
Omniverse, OpenUSD, Isaac
- 04Simulation
Physically accurate scenes
- 05Intelligent Robots
Policies that reach hardware
Research-led
Taught by people who ship, informed by people who research.
Resonae Labs is not a training company that added a research page. Learning is one of three outputs from the same team — alongside published research threads and delivered industrial systems.
That has a practical consequence for the curriculum: when a technique fails on a real plant floor, the material changes.
Research
Active threadsRobot learning, VLA, embodied AI. Open questions with prototypes attached, not a literature review.
Industry
Delivered systemsDigital twins, inspection cells, edge deployment. Systems with a customer, a deadline and a failure cost.
Learning
This divisionThis programme — the entry point into everything above. The confidential parts are stripped; the engineering is not.
What You'll Learn
Foundation of Physical AI and Robotics
Robotics Foundations
A working mental model of sensing, control and autonomy you can apply to any robot.
Physical AI Mental Model
A clear definition of Physical AI and how it differs from software-only AI.
NVIDIA Stack Map
An annotated map of Omniverse, OpenUSD, Isaac Sim and Isaac Lab and how they connect.
Isaac Sim & Isaac Lab Walkthrough
Hands-on, guided exposure to simulating a robot and watching a policy train.
Learning Path
How the eight days are shaped.
- 01Days 1–2
Robotics
The sense-decide-act loop, perception-to-action pipelines, and what separates automation from autonomy.
- 02Days 3–4
Physical AI
What makes intelligence "physical," and the simulate-train-validate-deploy loop it is built with.
- 03Days 5–6
NVIDIA Stack
Omniverse, OpenUSD and how the whole Physical AI stack fits together, layer by layer.
- 04Days 7–8
Isaac Sim & Lab
A guided walkthrough of simulating a robot and watching a policy train — then where to go next.
How Learning Works
Structured like engineering work, not like a course.
Live, not recorded
Eight daily sessions at 8:00 PM IST with the engineers who build these systems. Recordings exist for revision, but the room is where the questions get answered.
Eight days, one arc
Each day builds on the last — robotics, then Physical AI, then the NVIDIA stack, then a guided look at Isaac Sim and Isaac Lab.
Something every day
Each session ends with a short hands-on activity — a diagram, a comparison, a guided walkthrough — so the time is spent building understanding, not just watching.
No prior experience needed
No coding background, no GPU, no prior robotics knowledge. Built for a genuine beginner.
The Flywheel
Research → Industry → Learning
Research → Industry
Techniques leave the lab only when they hold under a customer's constraints: cycle time, safety, maintenance, cost of failure.
Industry → Learning
Curriculum is written from delivered systems. The labs are simplified versions of work that had a deadline attached.
Learning → Research
Teaching a technique to twenty engineers exposes every assumption in it. The questions that come back become research threads.
Upcoming Batch
Foundation of Physical AI and Robotics — Batch 02
Curriculum at a glance
What Is Robotics?
Perception, Control and Autonomy
What Is Physical AI?
The Physical AI Development Loop
The NVIDIA Physical AI Stack — Overview
OpenUSD and Omniverse
Isaac Sim — Simulating Robots
Isaac Lab and Where You Go Next
FAQ