Autonomous discovery engine

Startup ideas, found
by machines that never sleep.

Every 30 minutes, AI scouts sweep Hacker News, Reddit, Kickstarter, GitHub, Product Hunt, X, and founder chatter β€” then score what they find on creativity and real-world problem fit. Only the ideas that survive make it here.

Open the Dashboard β†’ … ideas ranked
…
Ideas tracked
…
Strong
…
Watch
…
Parked

How it works

A three-stage pipeline turns raw internet noise into ranked, evidence-backed opportunities.

STAGE 1
πŸ”­ Scout6 source agents sweep HN, Reddit, Kickstarter, GitHub, PH, X every 30 min
STAGE 2
🧹 GateDedup + taste filter. No real problem for real people β†’ parked, automatically
STAGE 3
πŸ”¬ AnalyzeCompetitor + market analysts dissect survivors: gaps, fit, GTM, scaling
STAGE 4
πŸ† ScoreWeighted 0–10 across 6 dimensions. 8.0+ pages a human immediately

Scoring rubric

Every idea carries its evidence β€” expand any card in the dashboard to see exactly why it scored what it did.

DimensionWeightWhat it measures
Problem severity25%How painful, for how many people?
Creativity / novelty20%Genuinely new, or a remix?
Market fit20%Reachable buyer with budget?
Competitive gap15%What exists β€” and where's the opening?
Scalability10%Grows 10Γ— without breaking?
Founder fit10%Plays to our strengths?

What makes the cut

The bar is deliberately high. Most of the internet doesn't clear it.

🎯

Real problems only

The creativity gate is ruthless: if an idea can't name a real problem for real people, it's parked. Cleverness without pain is noise.

βš”οΈ

Named competitors

Every survivor is mapped against actual incumbents β€” not "no competition" hand-waving. The gap has to be specific and defensible.

πŸš€

Built to scale

Ideas are scored on go-to-market paths, sales channels, and whether the model survives 10Γ— growth. Lifestyle businesses get parked.

Ready to explore?

Loading live stats…

Open the Dashboard β†’