AI scales judgment and expertise. It also masks (and eventually exposes) lack of judgment and expertise.
Read the issueThe argument only counts if I'm living it, so here is the tech stack I built.
The planner our homeschool runs on. A curriculum comes in as photographs of the book's pages, and the tracker schedules one lesson for each school day of the year. When a week doesn't go to plan, every date after it moves to fit. I built it with my wife using Claude Code.
What our sons learn, and in what order, stays with my wife and me. The tracker keeps the calendar.
Games I build with my sons to teach them coding and how to use AI without handing it their thinking: a sailing and fishing simulation on a real-time 3D ocean, a party game with its own physics and synthesized sound, and a robot game that teaches math and engineering.
The boys decide what each game should do, and test whether it does.
The thing that changed how I work the most: a coordinator, a builder and a reviewer. The builder proposes each change as a pull request, and the reviewer's job is to find what's wrong with it. A change merges only once it survives that criticism and I approve it. The coordinator is my line into both, so ambitious projects keep moving between my decisions.
Files #161: touch-size hit areas for the footer links.
Opens #164. Every footer link now has a tap area that meets WCAG 2.2. The footer looks exactly as before.
LGTM. Measured at 360, 390, 430, 768 and 1440 px: every hit area is 28.2 px tall.
Merge #162, then #164.
#164 needs main merged forward and a re-review first.
LGTM again. The merge is clean; nothing hand-resolved.
On Daniel’s word.
One Telegram bot out front and a team of Claude Code sessions behind it: a coordinator, health, legal/records, bookkeeping, family travel, and a pair that builds and reviews changes to the system itself.
Reading, drafting, briefs, summaries, plans, code review.
Merging a change, writing to the books, moving money, sending in my name, home security.
PlannedOne Telegram topic per subject, a private chat for approvals, levels from act to never with an audit trail, and an always-on home server.
Upload or paste in a résumé, and it scores how exposed the work is to automation and how ready the person doing it is to adapt.
It started as a memory platform for AI. Over time, I’ve evolved it into a context graph for multi-agent architectures.
I did this at scale for other people before I did it for myself: $2.5M saved at TaxAct, one account grown from $28K to $4M in nine months, $5M a year at NFL Digital Media.
Claude Code for Non-Coders is for senior technical people who built their careers on judgment, stopped writing code years ago, and are working out what AI and coding agents mean for them. You decide what to automate and what still needs your judgment. I build in the open and hand you a working version.
Tuesdaythe hands-on build, paid.Thursdayone idea in full, free.
Join 35,000+ senior technical people building their own AI agents and staying in control of the decisions.
Routes to Claude Code for Non-Coders on Substack. Unsubscribe anytime.
Step into the office and find the work. Everything in it is also on this page.
This device can't run the office. Everything in it is on this page.
A few companies a year bring me in on their specific problem: what to build, what to buy, where AI actually belongs in the operation, and how to keep their people the decision-makers instead of the assistants. I take the ones I can genuinely move.
I also give talks and run workshops: keynotes for leadership teams, hands-on sessions for the people who have to build the thing. If you want me in the room, start here.
I connect ideas and people that others see as separate, and I build what that connection makes possible.
I build my own tools, I write about how, and I advise the companies that want the same thing for their teams. I work in a window from early afternoon to evening, on purpose, because the mornings belong to my family, to fitness, to reading. I live on the coast of Virginia with my wife and two sons.
MS Tech Management, GMU · BS Computer Science, Bowie State · PMP · Lean Six Sigma