At Neuroinkx, I deployed a customer support chatbot on AWS EC2 using Python and FastAPI. I reduced its response latency by 40% through application and inference performance optimization, and implemented Docker containerization and CI/CD pipeline automation.
I'm building a multi-agent content generation system with LangGraph, including a planner–director–validator subgraph and persistent agent memory. I added an agentic RAG pipeline with GPTCache, human-in-the-loop approvals, and monitoring and evaluation with LangSmith, Opik, Sentry, and RAGAS.
