Case Study: AskBlue achieves rapid delivery of a location-aware recommendation engine in 4 weeks with ArangoDB

Creating a Location-Aware Recommendation Engine In 4 Weeks

AskBlue, a 150-person consulting firm working in finance and IT, built the location-aware BeOnit recommendation app for a startup client and faced a tight timeline and demanding technical needs: extremely fast, precise geo-based recommendations; storage and processing of GPS, Wi‑Fi and beacon data; graph-style relationships and native joins; and a horizontally scalable, open-source NoSQL solution that integrates with NodeJS. To meet these requirements they selected ArangoDB for its multi-model capabilities, AQL query language, geo support and cluster friendliness.

Using ArangoDB as the primary storage and processing engine, AskBlue ran proximity algorithms, campaign and notification logic, and complex subqueries directly in the database while connecting via NodeJS. The result was dramatic development speed and simplicity: a working prototype in two weeks and a public app launch in January 2018 (Android and iOS) delivered in four weeks, along with stable, fast queries, easier maintenance due to schemaless collections, and a system that scales for larger markets. ArangoDB enabled clearer, faster queries (via AQL), reduced development overhead, and supported production-ready performance and reliability.

AskBlue

Alex Pavlov

Solution Architect & Lead Software Engineer