redtachyon / autograd-hacks Goto Github PK

AI researcher at AI Redefined where I'm building tooling for human-in-the-loop RL. PhD from École Polytechnique, my thesis topic was "Simulating Crowds with Reinforcement Learning". In the past I was at University of Warsaw, KTH Royal Institute of Technology, and Aalto University.

In my free time, I'm the maintainer of Gymnasium (the official continuation of OpenAI Gym), sometimes reviewing other people's bugs, sometimes adding my own bugs... I mean, features. Sometimes I also help with PettingZoo, SuperSuit and other projects in the Farama Foundation.

My two main personal projects are Coltra and CrowdAI, which both form the core of my PhD research. Coltra is a somewhat general DRL library, designed to work seamlessly with multiagent environments and complex observation spaces. CrowdAI is a crowd simulation engine built on top of Unity ML-Agents in which I animate my crowds.

If you're interested in collaborating/hiring me/something of that sort, that might be a bit tricky at the moment. I'm open to contributions to Coltra, and ready to make it more community-driven, after a bit of extra code cleanup. Might be up for some freelancing, consulting and the like. And open-source contributions -- depending on my bandwidth. But if we already talked (e.g. on the RL Discord), chances are, I'll be up to help you with whatever you need.

Erdős number <= 4