Sanjay User

At Avasoft, I lead the design and delivery of AI-powered backend solutions across production projects, owning architecture, implementation, troubleshooting, performance tuning, production support, and Datadog observability for teams of 5–8 engineers.

I build multi-agent workflows on AWS Bedrock and Claude with tool calling, memory, retries, fallback logic, and Human-in-the-Loop approval gates. My customer-data automation reduced manual cleanup by 50–70% while ensuring zero unreviewed pipeline writes reached production.

I've delivered AI systems for healthcare, data mapping, manufacturing, and RAG search. These include a healthcare platform supporting 250+ patients and pharmacists, a data-mapping workflow that cut LLM calls by 20–30%, and a Smart Factory agent that reduced diagnosis time by 50%+.

I focus on grounded, guardrailed LLM systems end-to-end, using MCP/FastMCP tools, FastAPI, AWS, retrieval, evaluation pipelines, PII detection, and production monitoring.