I build production-grade LLM apps and agentic systems (Agno, LangChain, RAG) with strong MLOps on AWS/GCP.
- 🧠 Personal AI Assistant (full-stack; MCP/LangGraph/Agno) that connects to email, calendar, and files
- ⚙️ Production templates: FastAPI + async workers, observability, and K8s deploys
- 📚 Sharing notes on scaling RAG (caching, retrieval, evals)
- ML/LLM architecture & agentic workflows
- RAG performance (latency ↓, quality ↑), evals, and guardrails
- MLOps: CI/CD, containers, cloud, monitoring
Languages: Python · TypeScript
AI/LLM: Agno · LangChain · PyTorch · Transformers
Backend: FastAPI · gRPC · Async IO
Data/IR: PostgreSQL · MongoDB · Elastic · Neo4j
MLOps/Cloud: Kubernetes · Docker · Terraform · AWS/GCP · Triton · ONNX
- 🚀 Personal AI Assistant — FastAPI + LangGraph + MCP + EC2 deploy
- 📎 RAG Starter — Elastic / PG vector, evaluators, caching, guardrails (WIP)
- ✉️ Collaborations on production LLM systems & infra
- 🧩 Consulting: RAG performance, evals, and platformization