I design and optimize production-grade AI systems powered by LLMs. My focus is not building demos — it’s making AI systems reliable at scale. I’ve worked on: • Multi-tenant RAG platforms handling private enterprise data • Compliance-safe AI systems for government use cases • Agentic workflows using LangGraph with tool-use and memory • AWS-native architectures optimized for cost, latency, and scalability Common problems I solve: • RAG hallucinations caused by retrieval failures • Poor chunking and embedding strategies • High inference costs and latency bottlenecks • Scaling issues in multi-tenant AI systems If your AI system works in demos but breaks in production — that’s where I come in.