I am currently a PhD student exploring the intersection of Deep Reinforcement Learning, Physics, and Robotics. My core research focuses on bridging the gap between abstract AI reasoning and physical common sense.

I am actively building and looking for collaborators on projects involving:

• Physically-Grounded Reasoning: Teaching foundation models to understand causal world models through exploration].
• Vision-Language-Action (VLA) Models: Integrating multimodal inputs for robotic manipulation.
• Physics-Informed Neural Networks: Embedding physical laws into deep learning architectures for better sample efficiency.

I focus on tools that enable scalable simulation and real-world deployment.

Domain	Stack
Languages
DL & RL
Robotics
Edge AI	(Optimization & Deployment)

I am always open to discussing ideas or collaborating on papers/code in the following areas:

• Sim-to-Real Transfer: How to effectively transfer policies trained in MuJoCo/Isaac Gym to real hardware like the Franka Emika Panda or Hello Stretch 3.
• Edge Intelligence: Optimizing DNN training and inference on edge devices (Jetson Nano/Orin) — see my work on Pagoda and Fulcrum.
• Multi-Agent Systems: Experience with traffic control optimization using MARL.

Note to Contributors: If you are working on World Models or VLA architectures, I would love to connect and compare notes!

While I love code, I believe the best ideas come from interdisciplinary discussions. Hit me up if you want to talk about:

• Philosophy of Mind: Connectionism, Embodiment, and Reasoning.
• CS Education: How we teach complex AI concepts (I have experience mentoring and teaching).
• The Future of Robotics: From rigid manipulators to soft robotics and human-robot teaming.