Interactive textbook
Build a robot brain.No robot required.
You start from a blank simulation loop and build up, all the way to vision-language-action policies, your own physics engine, and a real robot arm. No framework to excavate, no black boxes. Free on a Colab T4 or your laptop.
No account, no setup, and your place is remembered locally in your browser.
See it work
live4096 robots learn at once
ch2.3You write parallel PPO on MJX. Here a 48-robot sample of the real run flails, then the whole field balances.
Now showing: 4096 robots learn at once. You write parallel PPO on MJX. Here a 48-robot sample of the real run flails, then the whole field balances.
Why this course is built the way it is
the code is the product
One file you can read top to bottom
Every chapter builds a single runnable script: a few hundred lines, no framework to excavate. You read it like a textbook, and every line you read is a line that runs.
runs on the free tier
A laptop or a free Colab T4 is enough
Every learner-facing path completes on CPU or a free T4. The wall-clock times on each page are measured on real hardware, never estimated; where a tier hasn't been measured yet, the page says so.
honest by construction
Real sims, real policies, real numbers
The browser demos run the same code you train locally. Seeds are mandatory, results reproduce within a recorded band, and every number you read traces back to a run, including the ones that show a method failing.
The arc
Six phases, one runnable file at a time.
43 chapters are live and readable today, across six phases, from a bare simulation loop to a from-scratch practitioner's stack and a real-arm graduation.
Phase 0 · Foundations
6 liveMake the simulator behave. Step physics by hand, author a scene, get frames and rotations right, and record your first teleoperated dataset.
Phase 1 · Imitation
9 liveTeach a policy from demonstrations. Behavior cloning, exactly why it breaks, and the models built to stop it from breaking.
Phase 2 · Reinforcement
8 liveLearn from reward instead of examples: for when demonstrations run out and the robot has to try, fail, and improve.
Phase 3 · Depth (optional)
9 liveOff the main line, taken when the itch strikes. Learn a world model, build a physics engine from scratch (dynamics, constraints, contact), compare it against the simulator you have trusted all along, then plan through your own engine with sampling-based MPC. The main line needs none of it.
Phase 4 · Post-Training
3 liveTake an already-trained policy and make it reliable. An offline-RL primer that learns from logged data, DAgger corrections that close the gaps demonstrations left, and human-in-the-loop RL (HIL-SERL) that turns a decent policy into a dependable one.
Phase 5 · Practitioner
8 liveBuild the practitioner's stack from scratch. A ViT and contrastive vision-language for perception, the two-tower VLA shape and a FAST action tokenizer, LoRA and INT8 quantization by hand, then the real-arm teleop → record → train → deploy loop that graduates you off the simulator.
Physical AI Weekly
The field, explained from first principles.
One mechanism teardown a week, plus a tight digest of what shipped. Honest, no hype.
Free. No spam. Unsubscribe anytime.