Senior AI Engineer with 11+ years of experience building production-grade AI systems, specializing in agentic AI architectures, LLM applications, and RAG pipelines. I help companies move from AI prototypes to scalable production systems.
Core expertise:
- Agentic AI systems using LangGraph, MCP (Model Context Protocol), and A2A protocols
- LLM application development with OpenAI, Anthropic Claude, Llama, and Groq
- Retrieval-Augmented Generation (RAG) pipelines with vector databases (Pinecone, FAISS)
- Fine-tuning and PEFT/LoRA for domain-specific models
- Production deployment on GCP (Cloud Run, BigQuery, GCS, Secret Manager, Firestore, Cloud Monitoring)
- LLM observability, evaluation frameworks, and multi-tenant SaaS architecture
Recent results delivered:
- 67% reduction in manual screening effort through agentic claims processing (2.8M+ claims)
- 28% improvement in clinical data extraction accuracy
- 65% lower error rate vs. classical ML baselines
- Precision improvement from ~40% to 85%+ on clinical RAG systems
I work best on greenfield AI builds, agentic system design, RAG architecture, and taking LLM POCs to production. Available for contract and consulting engagements.