Using Grab’s Trust Counter Service to Detect Fraud Successfully

Summary

Grab's Trust Platform team built the Counter service to detect fraud across business verticals like transportation, food, and payments. The platform replaces manual, multi-week engineering cycles with a self-service UI where data analysts can define and experiment with counters independently. Operating on an asynchronous ingestion and synchronous transaction model, the architecture evaluates incoming stream data, enriches it via internal services, and persists aggregated signals to ScyllaDB through Grab-Stats. A multi-bucket strategy partitions queries into fifteen-minute, hourly, and daily granularities to maintain low-latency query aggregations across wide time ranges under strict SLAs.

Takeaways

1. 1

   ScyllaDB was selected over in-memory stores like AWS ElastiCache because it was ten times cheaper, maintained equivalent stability, and achieved p99 read latencies under 150ms.

2. 2

   Pre-aggregating event metrics into 15-minute, 1-hour, and 1-day buckets enables scalable querying across diverse time ranges without performing heavy on-the-fly database aggregations.

3. 3

   The Counter service provides counter replay features across historical payloads to allow data scientists to evaluate and simulate new fraud features for machine learning pipelines.