Each case study covers the domain, the problem, and what shipped. Some were built alongside a team, some solo.
Agents read everything a portfolio produces, link it into a knowledge graph with full provenance, and surface conflicts, dependencies, and risks before anyone asks. Every claim traces to the exact region of the source page it came from.
An agent pipeline that produces a weekly client research briefing end to end. Every number is verified against primary sources, every chart is built from data filed in the repo, and each run records what it learned so the next one is faster.
Real-time AI research synthesis for hedge-fund rates PMs. It compares what markets, dealers, the Street, and the data say, then flags where they disagree.
AI swag allocation for a large events program: a Copilot Studio agent that proposes and negotiates packages within budget, paired with a dense admin app for the ops team, on one Dataverse backend.
An intake prototype where AI does the research and a person makes the decision. Requests come in through chat, and an agent assembles context by traversing Microsoft Graph.
Multi-tenant platform for Nurse Honor Guard chapters: public sites, member portals, and admin tools that scale without mixing data.
AI-powered fitness programming that builds custom workouts and long-term plans around your goals, gear, and history.