scraped every top ProductHunt launch from 2020-2025
27,000+ products. every tagline. every upvote. every category.
built something stupid with it:
→ click a button, get a startup idea
→ see exactly who you're competing with
→ find where they're weak
→ spot the gaps no one's filling
also found that saying "AI agent" gets +71% more upvotes than saying "simple" lmao
probably shouldn't be sharing this but fuck it
Claude, go find a SaaS to copy, undercut their price by 50%. While you’re at it steal their customers.
Use ultrathink, make no mistakes. I need to retire my bloodline in the next 24 hrs.
@personofswag “Let me guess you read Boris’s article on multi agent setups & Ralph + harness to enable 24hr of coding? Yeah yeah I know your whole fkn shtick”
There’s literally no more excuses
“I’m not technical enough”
“I don’t know if this idea is good”
“I don’t have a CTO”
Bro you got a fckn senior level engineer at your disposal 24/7, and they NEVER GET TIRED
Lock tf in
people don’t bounce off tools like claude because they’re bad.
they bounce because they don’t know where to put them in their life.
most people open it when they’re stuck. they ask a question. they get an answer. then they close the tab and go back to whatever they were doing before. next day, same thing. nothing carries over.
so of course it feels shallow. every interaction resets to zero.
what actually changes things is when you stop treating it like help and start treating it like a place.
same problem space. same type of work. same context, day after day. code, writing, planning, whatever your lane is.
over time it stops feeling impressive and starts feeling familiar. not smarter than you — just faster at the parts you already understand.
that’s when it becomes useful.
the leverage isn’t in better prompts or new features. it’s in staying in the same lane long enough for the tool to compound with you.