I design computational and engineering systems that connect neural activity to perception and behavior.
My work combines high-density BCI, multimodal imaging system, and machine learning models to understand and restore visual/motor functions.

At Rice University, I build end-to-end pipelines that record from the mouse LGN with 128-channel ultraflexible probes, design visual and microstimulation experiments, and develop neural encoding/decoding models that reconstruct visual information directly from population activity. I also create closed-loop behavioral systems, integrating deep learning tracking, real-time control, and neural decoders to study how the brain represents and transforms motor learning over time.

My long-term goal is to understand how light is encoded, perceive how it shapes neural systems, and build technologies that create new forms of light-driven experience.

Understand light. Feel light. Create light.