At SFLC.in, I’m designing end-to-end workflows and interactive prototypes for RightsLine, a legal-aid platform for cybercrime victims. I’ve partnered with engineers to validate feasibility and create reusable component patterns across product screens.
On my AI Video Assistant project, I built a meeting-intelligence pipeline that transcribes audio, summarizes meetings, and lets users query them through retrieval-augmented generation. I also built PaperLens to improve scientific retrieval by preserving figure context, resolving figure IDs, and grounding answers with citations.
For Bacterial Colony Dynamics, I led a team building an agent-based simulation with Double DQN, validated against published findings from the E. coli Long-Term Evolution Experiment. The project included a live visualization dashboard and Docker deployment.
I’ve also built projects in fraud detection, computer vision, and distributed systems, including a synthetic-identity fraud pipeline and an offline payment-routing system. I’m interested in applied machine learning, computer vision, NLP and retrieval-augmented generation, reinforcement learning for multi-agent systems, and secure distributed systems—with an emphasis on rigorous validation and reproducible pipelines.
