Senior Data Scientist, Detection

Cloudflare Cloudflare · Enterprise · New York, NY · Engineering

Cloudflare is seeking a Senior Data Scientist to research, design, and evaluate detection models for identifying automated, fraudulent, and malicious activity across Internet-scale data. The role involves analyzing massive datasets, defining success metrics for noisy ground truth, and staying current on AI/ML research, including LLMs and generative AI, to apply new techniques to their products. The position requires partnering with ML Engineers, Data Engineers, and Product teams to deploy detection approaches into production and measure their impact. Experience with fraud, abuse, and bot detection at scale, strong Python & SQL skills, and applied statistics/ML/AI fundamentals are essential.

What you'd actually do

  1. Research, design, and evaluate detection models that identify automated, fraudulent, and malicious activity across Internet-scale data.
  2. Dig into massive datasets to uncover the patterns and behaviors that distinguish adversaries from legitimate users.
  3. Define how detection success is measured, designing metrics and evaluation strategies for problems where ground truth is noisy, delayed, or contested.
  4. Stay current on emerging AI/ML research and evaluate how new techniques (e.g., LLMs, generative AI) can be applied to our products.
  5. Partner with ML Engineers, Data Engineers, and Product to take detection approaches from research to production and measure their real-world impact.

Skills

Required

  • Python
  • SQL
  • applied statistics
  • machine learning
  • AI methodology fundamentals
  • large-scale data analysis
  • fraud detection
  • abuse detection
  • bot detection

Nice to have

  • LLMs
  • generative AI
  • weak labels
  • delayed labels
  • absent labels
  • programming best practices
  • clear communication

What the JD emphasized

  • Fraud and bots at scale
  • ground truth ambiguity
  • detections when ground truth is scarce

Other signals

  • detection models
  • machine learning models
  • LLMs
  • generative AI