I’m Sumit Mishra (Ossm). I’m an independent researcher and engineer interested in Reinforcement Learning and Applied Mathematics.

I’m particularly interested in understanding how mathematical ideas such asprobability, optimization, dynamical systems, and mathematical modelingcan help us build better learning systems. I like learning things from first principles, implementing what I learn, and experimenting with ideas along the way.

Outside of research, I’m a competitive programmer with a1600+ rating on AtCoder, where I enjoy working on algorithms, data structures, and problem solving.

This website is where I share what I’m working on, includingresearch, technical writing, experiments, projects, and things I learn.

In the long run, I hope to do meaningful work at the intersection ofAI, mathematics, and science.