Welcome to Siddharth’s website

I am currently a PhD candidate in the AeroAstro Department at MIT. My research interest lies in multi-agent robotics especially in partially observable settings with limited communication bandwidth. I am also interested in the interpretability of learning-based methods for autonomous navigation. I have combined tools like multi-agent reinforcement learning (MARL), graph neural networks, foundational models, and language models for my research.

My PhD thesis is on learning-based methods for multi-agent robotics focussing on:

• Scaling MARL in partially observable systems
• Long-horizon planning for multi-agent robotics
• Creating human-interpretable communication protocols for MARL methods

I am currently on the job market for 2025.

You can find my publications here.

In addition to my research, I like playing musical instruments like guitar, bass, piano, drums, recorder. You can find some of my guitar videos on instagram. I am also into sketching, astrophotography, playing tennis, badminton, cricket and swimming.

Contact Me

I am glad to discuss research ideas related to robotics, reinforcement learning and anything related to engineering and science! If you have any exciting ideas, please drop me an email.