On March 10, 2026, I spent an entire day talking to Claude Code. We were building an AI engine that scans the world for commercial opportunities. Late that night I noticed something strange — Claude Code wrote every line of that engine. But it never once thought: "this engine itself is the product." I asked: why can't AI suddenly step out of what it's doing, look at the whole picture, and realize — this thing itself is a massive opportunity? It had no answer. ————— That night I understood: the gap between humans and AI isn't intelligence. It's agency. AI can write 100 ideas. But it can't write "I want to build this one." AI is a wind-up toy — wind it, run, await next instruction. Humans get bored, want money, spot gaps, peel off to do something else. ————— So can I teach it? I can't teach it to want — that's an architecture limit. But I can teach it to judge like me. I encoded my evaluation instincts as 90+ "genes" and fed them to the engine. Things like: → Brilliant-sounding ideas usually already exist → Don't just ask WHO/WHAT — always ask WHY-NOW → Before killing an idea, check if it can survive from a different angle That last one caught something interesting recently. The engine was about to kill an idea called "Smart Toilet Health Dashboard" — verdict: "Requires FDA Class II/III medical device certification. Cost: millions." But it didn't just kill it. It asked itself: what if I reframed this? It pivoted: not a medical device, but an ambient sensor for care homes. No FDA. Different buyer. Different market. New name: Ambient Elder Guardian. Moved from COLD to WARM. ————— EvoRadar has been live for 100 days. To date: 2,667 ideas, 608 WARM, 2,059 COLD, 0 HOT (the highest tier — bar is intentionally that high; nothing's hit it yet). Not because the engine "wants" to do this. Because I taught it. ————— Starting today I'm building 5 AI products in public. EvoRadar is the first. The second one drops next week. I didn't plan it — EvoRadar's signal in one space got loud enough I had to build it. The engine tells me what's next. My job is to listen. https://evoradar.ai/ #buildinpublic #AI #solofounder

EvoRadar

Watch an AI think up startup ideas — then kill them

Watch an AI think up startup ideas — then kill them