Building and maintaining real-time fraud detection systems presents significant challenges, particularly when detecting and preventing fraud at scale while maintaining millisecond-level response times. North tackled these challenges head-on by developing an advanced in-house ML system that could better adapt to emerging fraud patterns.

In this Q&A with Ben Orkin, Vice President of Engineering – MLOps at North, we’ll explore their journey from using a third-party solution to building their own fraud detection system powered by Tecton.

You’ll learn about:

• The business drivers and technical requirements behind North’s fraud detection model
• Key factors that influenced North’s build vs. buy decisions for their ML infrastructure
• How North improved their ability to iterate and adapt to new fraud patterns
• The future of fraud detection and opportunities for innovation

Whether you’re building fraud detection systems or working on real-time ML applications, this session offers valuable insights into architecting and scaling systems that demand both high accuracy and exceptional performance.
Join us to learn how North built a fraud detection system that can adapt to emerging threats while maintaining the strict performance requirements of a high-volume financial services platform.