IndieSignal

Real problems worth solving, curated from Reddit & HN

Visit Website
September 18, 2026 Turned my "what should I build" problem into a $9.99/mo SaaS

Hey IH, I'm sharing something I built solo over the last couple months (a good chunk of it with AI pair programming, not gonna hide that).

I kept doing the same thing every time I wanted a new side project: scroll "startup idea" threads for way too long, most of it recycled stuff with no real signal on whether the pain was actually there.

So I built IndieSignal. It scrapes Reddit, HN, Stack Overflow, GitHub Issues and Bluesky for people actually complaining about missing tools, then runs each complaint through GPT-4o-mini and scores it on pain intensity, existing competition, whether a solo dev could realistically build it, and regulatory risk (so you don't start building a fintech and find out later you needed a license).

It's $9.99/mo or $89.99/yr for the full catalog. Not because ideas should cost money, but scraping and LLM scoring across that many sources isn't free to run, and I'd rather charge a little than dump a huge list of stale junk on people.

Stack is Next.js, Supabase and Stripe on Vercel, with the scraping pipeline running on a Python cron. Wrote up the whole build log too, real bugs and real fixes, not a cleaned-up case study, link's on my profile if anyone wants that.

Live at indiesignal.app if you want to poke around. Happy to answer anything about the scoring or the stack.

I'm also working in some other ideas and I hope you like them.

Thanks for your time and attention.

Comment

About

I kept seeing the same complaints on Reddit and Hacker News, people who'd honestly pay to have a problem solved but nobody was building it. IndieSignal scans those posts and surfaces the ones worth building.