I work at the intersection of LLMs, RAG, AI agents, ML systems, and backend engineering.
I build and experiment with AI systems at the intersection of LLMs, RAG, AI agents, ML systems, and backend engineering.
Autonomous code-generation agent that turns natural-language goals into working Python projects.
Generation → Validation → Self-Healing → Recovery
Built with bounded LLM calls, automated pytest validation, traceback-driven repair, inference caching, and last-known-good fallback.
5+ merged pull requests contributing fixes and improvements to LlamaIndex, including framework-level fixes involving ReActAgent and OpenAI integrations.
Benchmarking Zero-Shot vs RAG code generation across multiple LLMs using functional test coverage as the evaluation metric.
AI: LLMs · RAG · AI Agents · Generative AI · NLP · ML
Engineering: Python · PyTorch · FastAPI · PostgreSQL · Docker · Git
Frameworks: LlamaIndex · LangChain
Building, experimenting, contributing to open source, and looking for opportunities to work on ambitious AI systems.