Fighting financial crime: How Kinectify uses Azure Cosmos DB, Fabric, and Azure AI to stop money...
summary
Kinectify builds a cloud SaaS that ingests thousands of gaming transactions per second and currently runs a transaction collection at 50,000 RU/s (doubling soon) using Azure Cosmos DB for geo‑replicated, single‑ms reads and multimaster writes. Pipeline uses Event Hubs → functions → Cosmos and streams raw data into Microsoft Fabric and a Medallion lakehouse for ML training. They ship a generic repository library, use Cosmos DB mirroring (preview) to sync into Fabric, and rely on Event Hubs to Fabric for near‑real‑time model retraining and anomaly detection.
AI-written summary · about
chapters (7)
- 0:00Introduction to Connectify
- 4:36Platform architecture
- 6:02Why Azure Cosmos DB
- 11:20Generic repository pattern
- 17:53Data ingestion pipeline
- 20:40AI and machine learning
- 25:09Lessons learned and conclusion
related talks
02mCustomer Story: Scaling Anti-Money Laundering Compliance with Azure Cosmos DB
17mAzure Cosmos DB in AI Workflows: From Knowledge Retrieval to Microsoft Fabric Integrations
20mThe What and the Why of Microsoft Fabric Real-time Analytics
29mInvestigate fraud with Azure Cosmos DB
46mDesign and build multi-tenant SaaS apps at scale with Azure Cosmos DB
01h 05mKeynote: Building Applications with the Modern SQL Using AI and Microsoft Fabric
46mIntegrating Azure AI and Fabric for Next-Gen AI Solutions
39mHow Microsoft Fabric Empowers Developers with Unified Data, DevOps, and AI