StaffDataScientist

ABOUT THE TEAM: Our Data Science team owns the machine learning backbone of Sift's fraud platform—a system that learns from 1T+ events annually across our network of 700+ global customers. You'll work alongside ML engineers, platform teams, and customer success leads who obsess over reducing false positives while catching sophisticated fraud patterns at scale. We're looking for a specialist who combines exceptional statistical rigor with deep fraud and information security domain expertise. You understand account takeover tactics, payment fraud vectors, identity manipulation, and network abuse patterns—not from reading threat reports, but from having modeled them in production. You'll be the go-to expert for diagnosing why models fail, architecting solutions across multiple modeling paradigms, and building processes that prevent data science from becoming a bottleneck. Your domain knowledge becomes a force multiplier: you'll spot feature opportunities others miss, anticipate how adversaries will probe your models, and translate customer fraud signals into modeling advantage. Success looks like: Models that outperform baseline by measurable margins because you engineered features informed by years of fraud pattern understanding. Production systems that don't degrade and don't leak money to evolving fraud schemes. Teams that trust your framework recommendations because you've debugged production failures in real fraud contexts. A research program that uncovers untapped signal in our customer data while staying ahead of attacker sophistication. WHAT YOU'LL DO: - Architect and own advanced modeling strategies across fraud and abuse problem domains (payment fraud, account takeover, identity spoofing, account abuse, content manipulation, credential stuffing). Your deep understanding of attacker tactics, exploit chains, and evasion strategies informs which signals matter and which are noise. You'll drive framework selection—deciding when gradient boosting on velocity featu