Selected AI work, built to run in production.

I build CGCircuit with a team of AI agents instead of one model doing everything. Each agent has a narrow job and its own focus, while one coordinator keeps them all working toward the goal. I gave a talk on this approach at Blizzard Cinematics.

• A team, not one big model. Each agent takes a role — planning, research, building, testing, security, review — and sees only what its job needs.
• A disciplined loop. Scope the work, put the right agents on it, build and test in parallel, review from several angles, refine until it's right, then ship and keep watch.
• Real work, in parallel. One agent can only do so much on its own; a coordinated team works on many pieces at once, with me directing the whole.
• The method is the point. Repeatable, reviewable, and built to last — the opposite of one-prompt guesswork that falls apart as the job grows.

An AI instructor built into CGCircuit's courses, so learners get help the moment they need it instead of scrubbing through hours of video for one answer.

• Answers in context. Learners ask questions against the course material and get sent to the exact moment that answers them.
• Search across the library. AI search over lessons and transcripts turns a video catalog into something you can query directly.
• Localized subtitles. Automatic subtitles in multiple languages widen who the courses can reach.

CGCircuit is a learning platform I designed, built, and still run — every layer of it. The storefront, the course-creation tools, and a custom video player people see; and the back end, the video processing, and the day-to-day running behind it.

• What people see. The storefront, tools for course creators, and a custom video player.
• The engine behind it. The back end that powers accounts, payments, live updates, and everything the platform runs on.
• Video that just plays. Raw uploads are processed automatically so they stream smoothly on any connection.
• Run, not just built. Growth, costs, and the day-to-day operations — I keep it running long after launch.

A workflow that turns messy input — raw meeting notes — into tracked execution: action items, a planned week, and a clear view of what got done.

• Notes in, actions out. Meeting notes are read automatically and turned into concrete, assignable to-dos.
• A planned week. Those to-dos are scheduled and prioritized into a realistic week rather than a growing list.
• Execution tracking. Progress is tracked against the plan, so what slipped is visible instead of forgotten.
• On the tools you already use. It runs quietly in the background on the stack the business already has.

Falcor is a personal “chief of staff” assistant, built on a reusable framework of my own — the pieces that turn a chat model into something that remembers, takes action, and runs on its own.

• It knows its job. Who it is, what it knows, how it works, and a memory that keeps what matters from one day to the next.
• Connected to the real world. It works with email, calendar, documents, and chat — and checks its own work so it doesn't quietly get things wrong.
• Runs on its own. It handles scheduled jobs — briefings, check-ins, reports — without being asked.
• Reusable by design. The same framework can set up a new assistant for a different role or business.