"We do not punish mistakes — we punish the absence of thought."

That's the guiding star of Project Axiom. It's also the thing I believe most deeply about education.

I've been teaching product management for several years now. And the thing I've learned — the thing that keeps surprising me — is that most students don't fail because they lack knowledge. They fail because they stop thinking. They reach for the pattern they've seen before. They answer the question they expected rather than the question that was asked. They confuse confidence with coherence.

Project Axiom is my attempt to build a world that makes that failure visible.

What Axiom is

Axiom is a PM simulator. Not a case study. Not a quiz. A world.

You play as a product manager at a fictional company navigating a crisis. Every decision you make has consequences that ripple forward. You can make the "wrong" call and still do well — if your reasoning holds together. You can make the "right" call and fail — if you can't explain why you made it.

The simulation doesn't grade you on outcomes. It grades you on coherence.

This is the core design principle, and it took me years to get to it. Early versions of Axiom were too outcome-focused. Students would game them: make the safe call, optimize for the rubric, produce answers that looked like the textbook. They were learning to perform PM work, not to do it.

The current version is harder to game because it doesn't tell you what "right" looks like. It asks you to build a consistent mental model and hold it under pressure.

The Six Pillars

Radical Continuity. Every decision you've made is visible to the simulation. You can't start fresh. Past choices constrain and enable future ones — exactly like real product work.

Cohesion over Correctness. There is no single right answer. The simulation evaluates whether your answers are internally consistent. You can take a contrarian position and do brilliantly — if you hold it with rigor.

Tax on Mental Laziness. Vague answers are penalized not with zero points but with confusion — the simulation proceeds as if you meant something specific, and you may not like what it inferred. This is how the real world works.

Fog of Strategy. You never have complete information. The simulation withholds things. You'll discover, mid-scenario, that you were missing context you didn't know you were missing. The skill being tested is not how you perform under full information — it's how you perform when you know you're uncertain.

Humanity of the Machine. The users in Axiom are people. They have names. They have anxieties and histories. The simulation surfaces their stories at unexpected moments. You cannot stay purely analytical.

The Legacy Audit. At the end of every scenario, the simulation runs a legacy audit: given every decision you made, what does this product say about what you value? This is the question most PMs never sit with.

Why I'm building it

I teach because I want to change how people think, not what they know. Most courses change what people know. The knowledge transfer is fine. But the habits of mind — the reflex to ask why, the discomfort with easy answers, the willingness to follow an argument to its uncomfortable conclusion — those are harder to teach in a lecture.

Axiom is my bet that you can teach habits of mind through a designed experience. That if you build a world where mental laziness has real, visible costs, people will develop the muscle to avoid it.

I don't know yet if the bet will pay off. Axiom is still in development. The first full scenario — a mental health app navigating a user safety crisis — is nearly done.

What I know is that the students who've gone through early versions come out different. Not because they learned new frameworks. Because they got caught. The simulation caught them pattern-matching when they should have been thinking. And getting caught — really, unambiguously caught, in a system that has no patience for excuses — is one of the most powerful learning experiences I know.

That's what I'm building.