Software Engineer, Account Abuse (Machine Learning)
Anthropic · San Francisco, CA | New York City, NY
About this role
**Software Engineer, Account Abuse (Machine Learning)** **About the Role** The Account Abuse team ensures Anthropic's computing capacity is allocated fairly and prevents bad actors from misusing resources. As a software engineer, you'll build machine learning systems to detect and stop abuse at scale. We're looking for full stack ML engineers experienced in model training, productionization, and evaluation. This is classical ML on structured and behavioral data—you don't need deep learning or LLM knowledge. What matters is shipping models where stakes are real and building robust production systems. False positives mean legitimate customers get locked out, so measurement, precision, and safe rollout are critical. **Key Responsibilities** • Build and operate a feature computation platform for training and real-time scoring with point-in-time correctness • Train, evaluate, and deploy models detecting account-level abuse and fraud • Automate model development lifecycle using Claude to speed up feature development and evaluation • Make backtesting, shadow deployment, and staged rollout standard practice with monitoring for skew and drift • Collaborate with data scientists and Policy & Enforcement teams on label quality • Partner with product teams to integrate model decisions with minimal latency impact **Minimum Qualifications** • Proficiency in Python and SQL • Experience training ML models and deploying to production • Experience with batch processing (Spark, Beam) and workflow schedulers (Airflow) • Understanding of point-in-time correctness and training/serving skew prevention • Strong communication skills **Preferred Qualifications** • Feature platform experience (Chronon, Feast, Tecton) • Stream processing engines (Flink, Beam, Kafka Streams) • Production ML with demanding serving requirements (fraud, risk, ranking) • Tree-based models on tabular data • Unsupervised/clustering/graph-based detection systems • Integrity, spam, fraud, or abuse detection experience • Experience with scarce, delayed, or noisy labels • AutoML or ML workflow automation • Interest in AI safety and societal impact **Compensation & Logistics** 💰 $320,000 – $485,000 USD annually 📍 Hybrid (25% office minimum) 🎓 Bachelor's degree or equivalent required 🛂 Visa sponsorship available *We encourage applications even if you don't meet every qualification. Anthropic is committed to diverse representation.*
Listing freshness
CronJobs last confirmed this listing 2h ago. If its source stops confirming the opening for seven days, this page is removed from active inventory.