Hello!

GitHub · LinkedIn · Twitter · E-mail

I'm Louis Le Lay, an incoming robotics research engineer. I'm working on making humanoid robots genuinely agile through reinforcement learning and simulation.

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

At PAL Robotics, I'm in charge of the RL locomotion stack for Kangaroo humanoid, from Isaac Lab/mjlab training to deployment with ROS/ROS 2.

Software

• mjlab

Project led by Kevin Zakka. Combines Isaac Lab's proven API with best-in-class MuJoCo physics to provide lightweight, modular abstractions for RL robotics research and sim-to-real deployment.

• Kinova Gen3 RL & Sim2Real Toolkit

Modular framework for training RL agents on the Kinova Gen3 robot using Isaac Lab, and deploying trained models either in Isaac Sim, or on the real robot via ROS2.

• SO-ARM RL: Manipulation tasks

Implementation of RL-based manipulation tasks for the SO‑ARM100 and SO‑ARM101 robots using Isaac Lab.

Talks

Training and Deploying RL Agents for Manipulation on the SO-ARM — Talk at ROSCon España, Nov 2025.

Open access

I keep most of my work public whenever possible.