I like software, from web development to artificial intelligence and statistical analysis. Recently, I have been especially interested in fast reinforcement learning: building high-performance simulators in C, C++, and CUDA so RL agents can learn from many more environment steps in less time. I have recently seen the power of RL (miss me with LLM RL) and wish to harness some of it. Here is an example from my twitter acount: laser-puzzle
- Building a fast conformer generation reinforcement learning environment conformer generation.
- Applying reinforcement learning to electric grid optimization.
- Making money on the stock market with an AI Trading Bot, albeit not very successfully.
- Building a key-value database from scratch in C: liteDB.
Check out my coding portfolio.
