Foundation · 1 of 3
The Context Audit: why your team's AI tools aren't landing
Your team has AI licenses and nobody's using them. That's not a training problem. It's a context problem.
Read the post →Start here
The teamvince build system.
If you're new to AI building, do not start with model comparisons. Start with the loop.
The operating loop
The shipping loop.
scope → spec → build → diff-review → verify → ship → write up. The same loop runs every project on the site.
- Scope the smallest real v1. Pick one user, one workflow, one cut line.
- Write the spec the agent can follow. AGENTS.md and CLAUDE.md, lightweight, no fluff.
- Let the agent build one change at a time. Small diffs beat big magic.
- Review the diff in plain English. If you can't explain it, don't merge it.
- Verify before trusting. Three checks every time: does it run, does it do the thing, did anything else break.
- Ship the tool. Internal or public. Real users, real data, real feedback.
- Capture the next-project brief. What broke, what worked, what's the next v1.
Seven steps. Repeat for every tool.
Best first reads
Three posts that set the foundation.
If you only read three things on this site, read these.
Foundation · 2 of 3
What I teach: a pattern from a dozen public builds
A dozen public builds, one repeatable workflow underneath. The map from the builds to what the course teaches.
Read the post →
Foundation · 3 of 3
How to write a lightweight CLAUDE.md (and AGENTS.md)
Anthropic's own guidance is to keep CLAUDE.md short. Bloated files drag every conversation down. The 2026 lightweight pattern.
Read the post →Best first action
Pressure-test the idea, then get the operating files.
The scope filter pressure-tests your idea. The harness gives you the operating files. Both are free.
01
Run the scope filter
Seven questions. Pick the lane, name the v1 cut line, copy a draft AGENTS.md you can commit. Five minutes, no email required.
Live tool
02
Get the harness
The agent harness Vince uses on his own projects. AGENTS.md, CLAUDE.md, five slash commands, plus an optional 60-second Loom take on your idea.
Free downloadWhat's next
Pick the path that matches the problem.
The harness gets you started. After that, the right next step depends on whether you want structure, private help, team delivery, or more reading.
Build with structure
Four live weeks, one shipped tool, a Skills library you own. Best if you want fixed dates, group momentum, and a finish line.
Build with private help
1:1 coaching for operators, PMs, consultants, and founders. Same workflow, your schedule, async help between calls.
Have your team build
Project-based consulting for messy workflows, AI prototypes, and agent systems that need to become usable.
Read the work
Short, opinionated essays on applied AI, agent workflows, and what makes Claude Code stick inside teams.
Start with the loop
Scope a v1. Get the harness.
Seven steps, repeated for every tool. Start by pressure-testing the idea, then drop the harness into a fresh repo and ship.
Run the scope filter