Building with real partners changed the kind of evidence I had.
A working prototype is one thing. Building alongside people who know the problem changes what the work means.
Read the field noteA public experiment by Alex
Can one person build a profitable software company with AI?
I’m testing that question in public by building real products, working toward paying customers and recurring revenue, and sharing what works, what fails, and what I change next.
Current status
A school booking and admissions platform is in staging, developed with industry partners.
$0Product MRR
15hCommitted weekly
StagingCurrent build stage
Real productsHonest numbersUseful failuresForward motion
What I’m building
The products I’m actively trying to prove.
Each active build has evidence behind it, a clear next test, and a reason to keep earning my time.
01Active
School booking and admissions platform
An AI-powered school tour booking and admissions platform.
- Evidence
- Developed with industry partners. The first progress review produced strong feedback that is now being implemented.
- Next proof
- Implement what we learned and return for the next progress review.
02Active
AI Dev OS
A system for planning, building, reviewing, testing, and continuing software work across AI coding agents.
- Evidence
- The CLI completed a small application lifecycle across Codex and Claude. DevHQ wireframes are complete and approved.
- Next proof
- Turn the approved DevHQ wireframes into a complete visual design.
The operating loop
Building is only one part of the job.
I have built plenty of things. The experiment now is whether I can repeatedly turn the right ideas into outcomes people value enough to pay for.
- 01Discover
- 02Validate
- 03Build
- 04Distribute
- 05Measure
- 06Learn
- 07Repeat
Field notes
The useful part is what actually happened.
The CLI proved the system. DevHQ is the next layer.
The underlying development system works through the command line. The next step is turning approved DevHQ wireframes into a designed interface.
Read the field noteWhy this matters
I’ve built for others. Now I’m building something of my own.
I’ve spent more than 15 years helping teams ship creative technology and software. What I have wanted for just as long is something of my own.
AI finally gives a product-minded builder like me the leverage to make that attempt real. This site is the record of whether consistent work, honest measurement, and better decisions can turn that leverage into freedom.
Read why I’m doing thisBuild Forward early access
Too many ideas? Decide what to build next.
Join the early access list for Build Forward, an AI project-decision skill that helps you choose what deserves attention, what to park, and the smallest useful test to run next. I’ll also share occasional updates from my own building journey, what I’m learning, and new skills as I release them.
Codex first. Claude Code and Grok Bot versions planned. Occasional emails. Unsubscribe anytime.