Everyone else agrees to avoid friction. Someone ends up with a disproportionate commute. Nobody mentions it.

The tool calculates each person's reachable zone using real transit data (not straight-line distance), finds the intersection, and surfaces Google Places results inside that overlap — ranked by a Fairness Score that measures how equally reachable a venue is for everyone in the group. Technical decisions worth discussing:

Isochrone-based rather than Euclidean midpoint — crow-flies distance is meaningless in a city with asymmetric transit

Per-person travel mode support — one person cycling, another on the Subway, a third driving

Session state serialised to URL hash so any search is shareable without accounts

Built entirely with AI tools as a non-technical founder — happy to talk about that process if useful

No accounts, no paywall, works in the browser.

Would genuinely appreciate feedback on the technical approach and any edge cases you'd throw at it.

hugpoint.io