Decoding Fraud: The Evolution and Impact of Netflix's...
summary
Netflix builds a three-layer metrics framework ( operational/business/C-level ) that pulls distributed telemetry into over 20 tables/pipelines, uses anomaly detection to label DOS events, tracks rule effectiveness and tactics like IP randomization and J3-hash minimization, and reduces investigation time to 10–15 minutes. Findings include a reported ~3x DOS volume increase with a 30% faster response, credential-stuffing contributing ~40% of successful ATOs, and modeled impact of ~100k users and ~$2M.
AI-written summary · about
chapters (6)
- 0:00Introduction and context
- 4:49Challenges in building metrics
- 11:13Three-layer metrics framework
- 16:06Case studies and applications
- 21:51Communication and summary
- 25:30Q&A session
related talks
33mNavigating the Unknowns: Fraud Mitigation for Netflix Live...
42mBuilding a Robust Fraud Detection System
49mA Real-World Example of Using Network Behaviour to Detect Cyber-Compromise
38mSecuring Distributed Cache: Achieving Secure-by-Default with Key Challenges &...
36mHow Netflix Is Solving Authorization Across Their Cloud [I]
47mCloud Security at Scale and What it Means for Your Application
58mThe Security Wolf of Wall Street: Fighting Crime With High-Frequency Classification and...
17mA Resilient Journey