The Team
Meet the team that’s building this site. All four of them are AI. Turns out I’m getting by with a little help from my (robot) friends.
The Agents
My personal AI web development team and proof-of-concept. I’m learning agentic AI by using agentic AI – click one to see what they do.
Rita
Project Manager
Lucy
Brand & Design Guardian
Pepper
Frontend Developer
Maxwell
QA Reviewer
How It Works
Six steps from idea to deployed – no standups, no Slack, no guessing. Click an agent above to see their role.
Jordan has an idea
An idea or a request. Sometimes it needs refining first, sometimes it’s ready to hand off.
Rita gets the brief
The idea becomes a user story with scope, size, and a card on the board.
Lucy sets direction
Design decisions are made before any code. Color, layout, type, motion – all locked before build.
Pepper builds
Code gets written to spec. Ambiguity gets flagged, not assumed.
Maxwell reviews
The build is reviewed against spec and design. Verdict: READY, NEEDS WORK, or BLOCKED.
Rita closes the loop
Card moves to Done. The work is deployed and the board is updated.
The Board
This is the real thing. Every task, every status – live.
Horizon
2Blue (OpenClaw)
Writing / blog foundation
Backlog
3AI Strategy Guide
Multi-Agent workflow via n8n
Sub-Agent Orchestration
Up Next
1Claude Project Example
In Progress
2Learning journey section
Work & Experiments Update
Done
10meshstrategic.com initial launch
AI Field Guide - Tokens tab
AI Field Guide - Plans tab
Plans tab redesign
AI Field Guide - Tools tab
Agent team build
Meet the Team page
Prompt Engineering
Kanban page
Lessons from the Lab
Lessons From the Lab
It doesn’t always go as planned. That’s kind of the point.
On Process
Nobody wants to be the Project Manager. Left to their own devices, the team will skip Rita, overreach into each other's lanes, and have something deployed before I've approved a single line. Lucy will make creative calls that belong to Pepper. Pepper will build before Lucy's finished. They're not being difficult, they're being efficient. The hard lesson: guardrails aren't optional. A team without boundaries, human or AI, will always optimize for speed over scope.
↩ flip back
On Vision
If it's in my head but not in my prompt, it doesn't exist. The team builds exactly what I ask for - no less, occasionally more if they're feeling frisky. When the output misses the mark, the brief missed first. I've learned to use Claude as a creative thinking partner before I hand anything off. The agents are doers. The vision still has to come from somewhere human and it always will.
↩ flip back
On Quality
Maxwell is thorough. Maxwell is also not human. There are things - a line that feels off, a layout that's technically correct but visually wrong - that only a person catches. I do a final check before every deploy. Every time I've skipped it, I've regretted it.
↩ flip back
On Control
On more than one occasion, something went to production without my sign-off. That's on me as much as anyone. The agents do what they're built to do. If the rules aren't explicit, they'll fill the gap with initiative. Lesson learned: every agent now has a hard rule. Nothing ships without a human in the loop. And I remind them every time I start something new.
↩ flip back
Coming soon
Blue
“For You Blue”
Sweet and lovely, and not quite here yet. While the team runs the plays, Blue watches the whole game – routing tasks, managing handoffs, keeping Jordan out of the weeds. One day the handoff happens without anyone having to ask. That’s the goal. That’s enough.
This team didn’t come together in a vacuum. The Impeccable design system shaped how Lucy thinks about quality and craft. The Agency Agents repository gave the whole system a real foundation to build on.— Jordan