Aakash Madabhushi
I train manipulation policies on a physical arm and build the evaluation apparatus that says honestly whether they work.
Graduate Research Assistant in the SJSU Department of Applied Data Science since March 2026, training and evaluating SO-101 pick-and-place policies on real hardware. The work covers the whole loop: teleoperated data collection, quality gates over the recorded episodes, PyTorch and CUDA training runs, and a physical evaluation harness that scores rollouts one at a time and survives being interrupted.
Training the policy is the easy half. The harder half is the measurement: a harness that returns the arm to start on its own, guards against stale caches, logs every failure mode separately, and reports 92% instead of rounding it up. That work now extends into Isaac Sim and MuJoCo for SO-ARM101 and Franka Emika Panda arms, so control approaches can be checked before they reach hardware.
Every link goes to code or data you can inspect yourself.
Scored 92% pick-and-place success and 2.49 cm mean placement error over 50 physical rollouts, by training an Action Chunking Transformer on 50 teleoperated demonstrations and scoring it against a protocol locked before evaluation.
Graduate Research Assistant, SJSU Applied Data Science · Mar 2026 to present
Takes physical robot time out of the iteration loop, by locating the real cube and bowl with ArUco markers from a calibrated overhead camera and mirroring their coordinates into Isaac Sim, so policies train against the real table's layout.
Personal project, built on NVIDIA's Sim-to-Real SO-101 workshop · Aug 2026 to present
Reduced robotics environment setup to one command on macOS and Linux, by packaging ROS 2 Jazzy, Gazebo Harmonic, RViz and TurtleBot3 into a version-pinned Docker Compose stack with a browser desktop that removes X11 failures.
Personal project · Mar 2026
Select a project for the full detail and links.
I'm available from December 2026 and open to relocating. Email me and I'll reply within a day.
Hiring for something else?