At Anviam Solutions, I designed, built, and evaluated production LLM agents for healthcare, including medical coding automation. I owned the evaluation layer, covering hallucination testing, token and cost optimization, and Speech-to-Text accuracy benchmarking.
I also architected clinical RAG pipelines with Neo4j, LangChain, and LlamaIndex, and built a Voice AI bot with automated escalation to a human agent. Earlier, I built LLM applications for a freelance client and worked on RAG chatbots, speech pipelines, translation models, and robotics systems.

