Louis Le Lay
simulation & rl for humanoids

I'm Louis Le Lay, a robotics engineer. At PAL Robotics, I owned the RL locomotion effort for the Kangaroo humanoid end-to-end: built the full pipeline from scratch and replaced the MPC-based walk the company had used for over 20 years. The stack is open source.

I also contributed to the creation of mjlab [ tweet ], an Isaac Lab-style RL framework built on MuJoCo Warp, and I'm an active contributor to Isaac Lab.

I'm working toward humanoid robots that move through the real world with the agility and robustness of humans.

GitHub · LinkedIn · Twitter · E-mail · Scholar

Louis Le Lay
Kangaroo humanoid walking on real hardware with RL locomotion
Kangaroo at 0.5 m/s — RL replacing 20 years of MPC
G1 robot getting up using reinforcement learning
G1 fall recovery
Tennis motion retargeting to a robot
Motion retargeting

Experience

PAL Robotics · Robotics Engineer Jul 2025 – Mar 2026

Owned the RL locomotion effort for the Kangaroo humanoid end-to-end. Built the full sim-to-real pipeline from scratch, led the technical direction, and replaced the legacy MPC-based walker with a learned policy. Olivier Stasse (Research Director, LAAS-CNRS) referenced my work on stage at VivaTech 2026 during a live humanoid demo.

University of Tokyo · Research Intern Jun – Sep 2024
Airbus · Robotics Engineer Intern Sep 2023 – Jun 2024

Publications

For a complete and up-to-date list, see my Google Scholar.

Kevin Zakka, Qiayuan Liao, Brent Yi, Louis Le Lay, Koushil Sreenath, Pieter Abbeel
arXiv preprint, Jan 2026

Software

PAL Robotics' robots in mjlab. Standard RL interface for Kangaroo, Talos, and TIAGo.
Isaac Lab API on MuJoCo Warp. Led by Kevin Zakka.
Isaac Lab training to real Kinova Gen3 deployment via ROS 2.
RL manipulation tasks for SO‑ARM100/101 in Isaac Lab.

Press & Mentions

Talks

Education