I build and stabilize production-grade AI backends — from RAG pipelines to voice AI systems. My core stack: FastAPI, LangChain, ChromaDB, PostgreSQL, Celery, Redis, and LLM integrations (Claude, GPT-4o, Retell AI).
What I offer:
- RAG engine development (document ingestion, chunking, vector search, retrieval tuning)
- LLM-powered backend APIs with FastAPI (async, scalable, production-ready)
- Multi-agent AI system architecture and orchestration
- Voice AI automation (Retell AI, LiveKit, real-time conversational agents)
- Production debugging: dependency conflicts, deployment issues (AWS EC2, DigitalOcean), Celery worker fixes, DB migrations
- AI system reliability and monitoring (Langfuse, Prometheus, Grafana)
Differentiator: I combine hands-on backend engineering with a business/marketing background (BBA), so I understand both the technical build and the commercial context — useful for client-facing MVPs, demos, and go-to-market-ready AI products, not just isolated code.
Recent work: built and deployed a HR voice automation system (Retell AI + GPT-4o + Kubernetes) for enterprise use, an AI audit platform with multi-agent RAG architecture, and stabilized multiple production AI systems handling real client traffic.
If your AI backend is broken, slow, or stuck in prototype mode, I can get it production-ready.