AI & Backend Engineer specializing in Agentic AI Orchestration, Production RAG Systems, and High-Concurrency Python Microservices.
- Agentic Workflows & Multi-Agent Systems: Designing cyclic state machines, reflection self-healing loops, and hierarchical supervisor patterns using LangGraph and FastAPI.
- Advanced RAG & Vector Retrieval: Architecting hybrid search (dense embeddings + BM25 sparse) with cross-encoder reranking and deterministic PII redaction on Qdrant.
- High-Throughput Streaming & Event-Driven Systems: Streaming, processing, and indexing high-volume transactional and transcript data using Apache Kafka, Redis, and PostgreSQL.
- Deterministic Guardrails & Model Serving: Bridging frontier LLMs with private open-weight code models via vLLM, enforcing deterministic AST security policies.
| Domain | Technologies |
|---|---|
| Languages | Python (AsyncIO), TypeScript, SQL |
| GenAI & Orchestration | LangGraph, LangChain, AWS Bedrock, vLLM, Model Context Protocol (MCP), Ragas, LangSmith |
| Vector DBs & Search | Qdrant, Hybrid Search (BM25 + Dense Vectors), Cross-Encoder Reranking |
| Backend & Messaging | FastAPI, Apache Kafka, Redis, Uvicorn, Docker |
| Databases & Storage | PostgreSQL, MySQL, MongoDB, DuckDB |
