AI systems builder & educator

AI agents can now build software from a written plan. So the person who writes the plan decides what gets built, and in most teams that is the product manager or the product owner. However, every decision the plan leaves out, the agent makes for you, and the team finds it in review or in production.
Spec-driven development fixes the order: decide first, write it down, then let the agent build from it. You start by hand with Claude Code, so you see the loop. Then you run the BMAD Method, which is free and open source: it interviews you like a product team, splits your idea into stories, builds each one, reviews it and looks back at the end.
You do it both ways. Greenfield, which means a new product built from an empty folder. Brownfield, which means code your team already runs and changes every week.
Developers and tech leads are welcome too. Every step runs on real code in Claude Code, and it is the same loop a team uses to keep AI-written code to what was decided.
I have spent 20 years building software platforms for enterprises. The whole course is one recorded run, and the kit has 22 saved states. So you can repeat it step by step, and rejoin at the next lesson if your run drifts.
Go from writing tickets for developers to writing specs AI agents build from, on new products and on code your team already runs.
Spot the five problems vibe coding creates, and trace each bug back to a decision the AI made for you.
Organise a project around foundation files and one spec per feature, approved before any code exists.
Turn a prompt you type twice into a Claude Code skill, so the loop repeats without retyping it.
Pressure-test a one-paragraph idea with BMAD, one question at a time, so the rules nobody wrote down get decided.
Write a five-field spec (why, capabilities, constraints, non-goals, success signals) with every open question closed.
Decide when a PRD is worth writing, using the test from BMAD's own docs.
Order stories so each needs only the ones before it, the way a good backlog reads.
Pin shared decisions (stack, database, data shape) in an architecture spine before any code exists.
Run every build with a six-line checklist, and read what the agent flags before you approve it.
Sort review findings into fix now, defer or decide, and catch the repeated finding that points to a gap upstream.
Run a thorough review and a retrospective that checks the whole release against the spec.
Prove the product with test days you chose before running them, and leave with a report your team can read.
Size the work first: skip the planning, send a small change straight to a build, or plan a full epic.
Ship a one-sentence change in existing code, then read the diff for changes nobody asked for.
Build a new feature from a spec that read the existing code first, and record what the code cannot say.

AI systems builder & educator. 20 yrs enterprise software, 20K+ audience
Product managers and product owners whose job is now the spec, and who want AI agents to build from it instead of guessing.
Technical founders building their own product with AI, who need it to hold up past the demo, feature after feature.
Tech leads, developers and tech teams who want AI-written code to match what the team decided, in new and existing code.
Every lesson runs in Claude Code. My full run used a Max plan with no limit hit; on Pro, expect to pause between stories.
No code to write. The lessons show a terminal. In the Claude Code desktop app, Claude runs the same commands for you.
You install Node.js, Git and uv. Some work laptops block installs, so a personal laptop can be easier.
25 video lessons, about 4 hours 45 minutes
Five modules. First spec-driven development by hand with Claude Code, then the BMAD Method on a greenfield product, from a rough idea to a reviewed release. Then BMAD on brownfield code. A bonus lesson comparing BMAD with Spec Kit, OpenSpec and plan mode is coming soon.
The course kit, with 22 saved states
The starting files, the finished app and a saved state at every step of my run. If your run drifts, load the state and rejoin where the lesson starts.
5 hands-on exercises
One per module, on your own copy of the kit or your own project. Each has a "done when" list, so you know when the module is finished.
Monthly live group call
One live call a month with me for everyone enrolled. Bring a spec, a stuck story or your own codebase. Recordings stay in the course.
Lifetime access
Keep the lessons, the kit, every call recording and the comparison lesson when it lands. When a BMAD release changes a step, the lesson notes say what changed.
Certificate of completion
Finish the five exercises and get a Maven certificate to share on LinkedIn or with your employer.
Maven Guarantee
Your purchase is backed by the Maven Guarantee.
Stage 1. See the loop by hand. You write a spec for a small product in Claude Code, let the AI build from it, and see why a vague prompt goes wrong.
Stage 2. Turn your idea into a spec. BMAD interviews you one question at a time, until the rules nobody wrote down are decided.
Stage 3. Plan it like a product team. The spec becomes ordered stories and one page of shared decisions, before any code exists.
Stage 4. Get it built and checked. Each story is built in a fresh chat, reviewed, and proven with test days you chose up front.
Stage 5. Do it on brownfield code. The same loop goes into an existing codebase: a small change first, then a full feature.
Where you end up: you own the decisions, the agents do the building, and you can check the result yourself. That is what working AI-native looks like.
Maven for Teams
Reimbursement
Get your company to pay
Everything L&D needs: email template, receipts, and certificate of completion.
Get reimbursedPrivate cohort
Run a cohort for your org
A dedicated cohort with a custom schedule and curriculum, tailored to your team.
Book a private cohort