Built a RAG system, streaming ingestion pipeline, and custom vLLM-based LLM inference service on EKS on AWS GovCloud for Tesla Government, a contracting firm that provides information management and data analytics services to the State Department and other government agencies.