At Databricks, I built a GenAI observability and monitoring platform to track LLM performance, latency, token usage, and errors in production. I also developed RAG pipelines and evaluation and prompt optimization workflows, improving response quality and reducing token usage costs by 20%.
Previously, at Anthropic, I developed AI-powered applications integrating Claude LLM APIs, and at ProductiveEdge I built Go-based backend systems and React features. Across these roles, I’ve worked with APIs, data pipelines, and cloud deployments to improve application performance and reliability.
