Dev preview · 400k POIs across 50 cities

Points-of-Interest ranking,
tuned to your customer.

A single REST API over OpenStreetMap + Wikidata + Wikipedia pageviews. Each customer picks their own tier weights, category boosts, and distance decay — same dataset, completely different rankings.

City
Profile

Top 5 near Paris

  1. No results.

cURL

Updates as you change inputs
curl 'https://api.poiscore.com/v1/pois/nearby?lat=48.8584&lng=2.2945&radius_m=2000&limit=5&tier_s=5&tier_a=3&tier_b=1.5&tier_c=0.8&base_offset=0.5&w_fame=0.3&w_quality=0.2&tau_m=800' \
  -H 'Authorization: Bearer sk_live_...'

Live API calls to /v1/pois/nearby — flip a profile chip and watch the ranking change.

0
POIs ingested
0
Cities live
0
Wikidata-linked
0
Wikipedia articles

What makes it different

⚙️

Tunable scoring

Per-request weights + category boosts + distance decay. The same query at the Eiffel returns Tour Eiffel #1 by default, or museums #1-5 for a family travel app.

€0

Free upstream data

OpenStreetMap, Wikidata sitelinks, Wikimedia pageviews. Zero recurring data costs versus Google Places at scale.

Fame, not ratings

Wikidata sitelinks + 30-day pageviews build a deterministic fame_score. No fake reviews, no user content moderation.

PostGIS-powered

GiST index + ST_DWithin sub-100ms over 165k POIs locally.

🔑

First-class API keys

sk_live_* Bearer tokens, scrypt-hashed at rest, daily rate limits, per-key usage metering.

🇪🇺

EU-native, self-hosted

Postgres + Redis + Caddy on a €5/month Hetzner VPS. Customers can on-prem the whole stack.

In this preview