Heyy! I am Ishaan, working towards post-training LLMs, learning RL and tryna squeeze FLOPs Currently improving multi-step reasoning in agents via RL @ https://www.atomicwork.com/ Prev: post-training @ https://www.sarvam.ai/ alum @ https://www.iitr.ac.in/