As AI-Driven Developers, we are discovering better ways of building software by working with AI as a deliberate partner — and helping others do the same.
Each value below is a compass, not a rule. Together they map the four places a developer's value now lives — method, ownership, understanding, impact. Read the line. Feel which side the weight falls on — then notice that the other side never disappears. It simply weighs less.
Four tensions. Each side still matters. The weight falls left.
plan ──→ implement ↑ ↓ assert ←── review ——————————————— any ide · any model workflow leads.
$ claude $ codex $ cursor $ copilot … swap tools · same speed.
$ git commit → reviewed ✓ → tested ✓ Signed-off-by: you ✓ ——————————————— the human owns it.
$ ai "ship it" + 1,240 lines merged ——————————————— ⚠ no review nobody read it. who owns this?
ai: auth via JWT. $ why? we use HttpOnly sessions. ai: rewriting. ✓ ——————————————— you understood.
ai: auth via JWT. $ ok accept ——————————————— ⚠ 3am why is this here? nobody knows.
features shipped ✓ bugs avoided ✓ users impacted +410 ——————————————— the business moved.
lines generated 12k tokens burned 1.2M commits/day 38 ——————————————— … vanity · zero impact.
That is, while there is value in the items on the right, we value the items on the left more.
Read them slowly. Argue with them. The document is stronger for every sharpening.
We deliver working software to production, faster and without debt.
Speed and quality are not in conflict. They are the same discipline, practiced twice.
Planning is our first act of collaboration with AI.
The quality of our plan determines the quality of our output. Everything downstream inherits it.
We break complexity down until it can be delegated cleanly.
Clarity of scope is clarity of output. Ambiguity compounds; decomposition dissolves it.
Context is our primary asset.
We invest in it, maintain it, and protect it from noise. A polluted context produces polluted code.
We apply the same standards to AI-generated code as to any code we write by hand.
There is no lower bar for generated work. Review, test, reject — the same rituals apply.
We are the architects. Every line that ships is our line — regardless of what produced it.
Authorship is not about the keystroke. It is about the judgment that sent the keystroke.
We use AI to go further in what we already understand.
We do not delegate what we have not mastered. Competence first; acceleration second.
When a session loses coherence, we restart.
A clean context is always cheaper than a compacted one. Fresh starts are not failures — they are hygiene.
We choose a consistent methodology over a collection of tools.
The method outlasts any model. Tools rotate; practice accrues.
What one of us discovers, all of us adopt.
The community is our compound learning machine. Hoarded technique decays; shared technique compounds.
We recognize AI's limits as signals — not walls to push through, but invitations to decompose further.
When the model stalls, the prompt is wrong. When the prompt is wrong, the scope is wrong.
We do not stop learning because AI can code. We learn faster because it can.
The leverage is real. What it multiplies is up to us.
One YAML file, one pull request, on GitHub. No backend, no form, no email harvest. Your name lives next to the manifesto, in public.
Five lines to fill, reviewed within a few days. The guide walks through it.
This is how I already build software with AI, and how I want the trade to mature.
AI-Driven Ops
AI-Driven Development. Craft first, always.
AI-Driven Development, create things and have fun with it
AI-Driven Development, the method that helps me to keep control of my codebase quality in an AI driven world.
AI-Driven Dev
Master AI for speed and excellent products that people love
AI is not a trend, it's how we work. AI-Driven Dev is the path I chose to build better, faster.
