Build and deploy production LLM/ML pipelines for patent workflows: retrieval, multi-step prompt pipelines, evals frameworks, fine-tuning, agentic workflows, monitoring, and cloud deployment. Collaborate with attorneys and cross-functional teams to convert domain expertise into robust, scalable inference and data solutions.

Design, build, and maintain end-to-end full‑stack features across Python/FastAPI backend, Next.js frontend, and LLM inference pipelines. Collaborate cross‑functionally to deliver scalable AI-powered patent analysis, optimize performance, address production issues, and reduce technical debt while improving engineering practices.