Case Study: Apartment List achieves real-time personalized apartment recommendations with Chalk

How Apartment List uses Chalk to power personalized apartment recommendations

Apartment List, an online marketplace for apartment hunting, faced significant challenges in delivering real-time, personalized search recommendations to its users. Their previous system, which relied on custom-built pipelines and manually managed feature data from multiple sources, resulted in delayed search updates, duplicated engineering efforts, and performance bottlenecks. These issues hampered the user experience as renters expected instant results when refining their search criteria. They turned to the vendor Chalk and its real-time feature platform to overcome these obstacles.

By implementing Chalk's platform, Apartment List unified data from SQL, streams, and APIs to compute features in real time. The solution enabled instant search updates, dynamic personalization based on user behavior, and sub-5ms query response times. This resulted in a more responsive and intuitive user experience. Furthermore, Chalk drastically improved engineering efficiency, reducing the time to deploy new ML models from weeks to just days and eliminating data latency issues.

Apartment List

Matt Weale

Software Engineer