Focused on AI agents, LLM applications, retrieval systems, document intelligence, and scalable ML infrastructure.
I enjoy building systems that go beyond simple LLM wrappers — multi-step agents, tool execution, structured extraction, retrieval, evaluation, guardrails, and reliable backend infrastructure.
- 🤖 AI Agents — orchestration, tools, planning, memory, human-in-the-loop workflows
- 🔎 RAG & Deep Research — heterogeneous materials, semantic retrieval, context construction
- 📄 Document Intelligence — parsing, extraction, normalization, validation, cross-document reasoning
- 🧠 LLM / Foundation Models — inference, evaluation, multilingual & multimodal systems
- ⚙️ AI Infrastructure — model serving, async pipelines, observability, distributed workloads
- 🔗 Distributed Systems — storage, state management, blockchain infrastructure
Most repositories here explore different parts of the same problem:
how to build autonomous AI systems that can reliably understand information, reason over it, use tools, and execute real work.
Recent work includes:
- agentic research and knowledge systems
- LLM routing, inference, and evaluation
- coding / computer-use agents
- multimodal agent evaluation
- sandboxed tool execution
- retrieval and document-processing infrastructure
- distributed storage and blockchain systems
Python · Go · Rust · TypeScript · PyTorch · FastAPI
PostgreSQL · pgvector · Redis · Docker · Kubernetes · AWS
Research → systems → production.