Applied AI Engineering

The full AI engineering stack from end to end — RAG pipelines, autonomous agents, evaluation frameworks, and production deployment.

The curriculum covers seven weeks of hands-on modules: building and evaluating RAG pipelines, designing autonomous agents with LangChain and LangGraph, implementing vector databases and Graph RAG, building SQL agents with MCP integrations and memory systems, and working with modern patterns including evals, fine tuning, and spec-driven development. Every module ships working code — not slides.

The program culminates in a two-week on-site intensive where teams build and deploy a functional AI system for a capstone project focused on real business ROI, followed by a Demo Day presentation.

I served as both product manager and engineer on a three-person team — bridging product thinking with hands-on technical execution.

On the product side, I owned the roadmap, requirements, and cross-functional delivery. On the engineering side, I designed the multi-agent architecture for our capstone — a customer-facing AI system that turns natural-language questions into validated SQL queries against a healthcare SaaS database — and built the end-to-end evaluation harness covering intent classification, SQL correctness, security boundary enforcement, and response quality.

The biggest shift for me was evaluation rigor. Learning to build systematic evals — not just vibes-checking outputs — changed how I think about shipping AI. I'm now applying that same discipline to AI-assisted design workflows and AI product features in my current role.