Study list, clubbed by project
Named videos where we have a stable ID. Channels and search queries fill the rest. One scroll per project, not a scavenger hunt.
MuJoCo foundations
Physics and control on the laptop: hand-written PD, a scripted fish gait, then a small PPO swimmer. Everything lands in the harness.
Isaac Lab factory
Train a Franka arm to lift a cube with RL at 4096 parallel environments, and measure what rendering actually costs.





System ID on the arm
My first sim-to-real project. Run the same motion in MuJoCo and on the real SO-ARM101, then tune the simulator until it matches the arm on held-out trajectories.
Imitation learning on the arm (tethered)
50 demos collected on the arm, trained on the 5090, policy runs on the workstation with the arm on a cable. P5 cuts the cord.
Edge deployment
Unplug the 5090 (unplug the workstation). The arm keeps doing the task. The exact P4 policy now runs untethered on a $249 Jetson at 15 W.
Autonomous rover + cloud skill loop
A voice-commanded Yahboom M3 Pro that navigates locally, plans with Claude, remembers through AWS, and closes its sim-to-real gap with readable correction skills.
Tennis-ball collector
The P6 rover plus a detector plus an intake. A working Tennibot, built so you know which parts were actually hard.
Robot menagerie
Point common/eval_harness, sysid, and deploy at robots you did not design. Or find out they were three scripts in a trenchcoat.
Legged sim-to-real
Train locomotion at scale, ablate domain randomization on hardware, and attribute the gap to actuator limits you already measured.
Whole-body / VLA
Fine-tune a VLA on the arm you own, compare it to P4 ACT, and do not fake a humanoid.

