Re:Ideavo – Tinder-style validation for startup ideas
by ideavo on Monday December 08, 2025 @12:23AM
(#46188688)
Hello HN,
I'm a solo dev/founder from India. I built Ideavo to solve the "Cold Start" problem for validating early-stage startup concepts.
The Problem: On most platforms (Reddit/Product Hunt), validation is a popularity contest. If you don't bring an existing audience, your idea gets buried immediately. There is no algorithmic fairness for new, raw concepts.
The Solution: I built a swipe-based feed (React + Supabase) that forces a "fairness queue."
Guaranteed Impressions: Every new idea gets a guaranteed "test flight" number of views in the feed before the ranking algorithm takes over.
Anti-Spam: I added a speed limit on the swipe interaction to prevent click-farming; users must pause briefly to read the card.
The Stack: Built with React, Supabase (Auth + RLS), and Edge Functions for the ranking logic.
We currently have ~150 ideas seeded from my local community in Gurgaon. I'm trying to figure out if this mechanics-based approach yields better signal-to-noise than standard upvote lists.
The UI is still in MVP state. Would love feedback on the ranking logic or the interaction model.
I'm a solo dev/founder from India. I built Ideavo to solve the "Cold Start" problem for validating early-stage startup concepts.
The Problem: On most platforms (Reddit/Product Hunt), validation is a popularity contest. If you don't bring an existing audience, your idea gets buried immediately. There is no algorithmic fairness for new, raw concepts.
The Solution: I built a swipe-based feed (React + Supabase) that forces a "fairness queue."
Guaranteed Impressions: Every new idea gets a guaranteed "test flight" number of views in the feed before the ranking algorithm takes over.
Anti-Spam: I added a speed limit on the swipe interaction to prevent click-farming; users must pause briefly to read the card.
The Stack: Built with React, Supabase (Auth + RLS), and Edge Functions for the ranking logic.
We currently have ~150 ideas seeded from my local community in Gurgaon. I'm trying to figure out if this mechanics-based approach yields better signal-to-noise than standard upvote lists.
The UI is still in MVP state. Would love feedback on the ranking logic or the interaction model.