Sam Daneshvar

Machine Learning Engineer based in San Francisco with 6+ years of experience building and scaling production ML and LLM systems. Specializes in agentic AI architectures, RAG pipelines, inference optimization, and safe large-scale deployment. Led multi-agent workflows using LangGraph, cutting inference latency by 50%, reducing LLM costs by 35%, and improving retrieval accuracy by 45%. Built LLM evaluation frameworks with automated regression testing and CI/CD-integrated A/B testing. Published researcher with 90+ citations.