Sutharsika Kumar

Hey, I’m Sutharsika, and I love trying new things. Naturally, that has led me to explore various hobbies and topics. This is a small collection of my work.

I’m currently working on reinforcement learning policy, when language models can serve as policy optimizers, and on developing an autonomous robotic laboratory, advised by Dr. Haozhe Wang at Duke. The lab work builds on Co-Scientist, a Gemini-based system that plans experiments and drives real laboratory equipment in closed loop. Previously, I collaborated with researchers at Duke, MIT, and Google DeepMind on ATOMIC, a zero-shot autonomous microscopy framework for 2D materials. Feel free to reach out at sutharsika.kumar07@gmail.com.