_

Agentic Research Paper Reader is a multi-agent research assistant that understands natural-language research queries, searches arXiv + OpenAlex + Crossref + Tavily, stores papers in Supabase + pgvector, ranks and reflects on results, and lets you select papers for deeper analysis and insight synthesis. The system is built with FastAPI, uses DeepSeek for reasoning (and optional embeddings), and includes a small HTML/JS frontend.

Developed "Headnote LLM", a domain-specific 8B-parameter model by fine-tuning Meta Llama-3.1 on 10k+ legal judgment datasets. Achieved state-of-the-art performance on legal reasoning tasks while adding only ∼168 MB of trainable parameters via LoRA adapters.

StatArb is a comprehensive Python platform for statistical arbitrage across equities, FX, crypto, and metals. It combines multivariate basket stat-arb (GMM-HMM regime switching), factor models, optional XGBoost signals, and advanced risk management/analytics into a single, configurable trading and research toolkit.