MS CS @ USC (graduating May 2026). Previously MLE at an early-stage startup in India.
I work on machine learning systems — graph neural networks, retrieval-augmented generation, and applied ML for healthcare.
- Under review at Knowledge-Based Systems (Elsevier) — Multi-turn glaucoma patient chatbot - "GlaucomaGuide"
- RESL Lab, USC — Multi-Agent RL for non-prehensile robotic manipulation
- InfoLab, USC — Dementia Precision Risk Tool
- Graph RAG Nutrition Chatbot — End-to-end Graph RAG pipeline over 30 nutrition papers. Leiden community detection, hybrid retrieval (entity + semantic), Pinecone + Llama3-70B.
- ICGRNN — Input convex graph recurrent neural networks for heat diffusion control on graphs. Convexity-preserving message passing + temporal updates.