SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Salary: USD 320,000 - 485,000 / annual
Anthropic is seeking a Staff+ Software Engineer to join the Account Abuse team, which is responsible for detecting and preventing abuse of Anthropic's computing capacity at scale. This is a full-stack machine learning role focused on building production systems that identify and stop bad actors while minimizing false positives that could lock out legitimate customers.
You will own the end-to-end ML lifecycle: building feature computation platforms that serve both training and real-time scoring with point-in-time correctness; training, evaluating, and deploying models for account-level abuse and fraud detection; and creating tooling to automate model development workflows using Claude. You'll establish best practices around backtesting, shadow deployment, and staged rollouts with comprehensive monitoring for training/serving skew, drift, and adversarial adaptation.
This is classical ML on structured and behavioral data—not deep learning or LLM internals. The work emphasizes robust production systems as much as model quality. You'll partner with data scientists, policy & enforcement teams, and product/platform teams to gather signals and integrate model decisions with minimal latency or architectural impact.
Anthropics's mission is to create reliable, interpretable, and steerable AI systems. The company is a quickly growing group of researchers, engineers, policy experts, and business leaders working on beneficial AI. They value impact, empirical rigor, and collaboration, with frequent research discussions to ensure high-impact work.
The role is hybrid with an expectation of at least 25% in-office time at San Francisco or New York offices, though some roles may require more.
REQUIREMENTS:
Minimum qualifications:
- Proficiency in Python and SQL
- Experience training machine learning models and deploying them to production
- Experience building data pipelines with batch processing engines (e.g., Spark, Beam) and workflow schedulers (e.g., Airflow)
- Working understanding of point-in-time correctness and training/serving skew prevention
- Strong communication skills and ability to explain technical tradeoffs to non-technical stakeholders
- Bachelor's degree or equivalent combination of education, training, and professional experience in a field relevant to the role
Preferred qualifications:
- Experience building or operating feature platforms (Chronon, Feast, Tecton)
- Experience with stream processing engines (Flink, Beam/Dataflow, Kafka Streams)
- Experience training ML models in production with demanding serving requirements (fraud, risk, ranking)
- Experience with tree-based models on tabular data
- Experience building unsupervised, clustering-based, or graph-based detection systems for coordinated account abuse
- Experience in integrity, spam, fraud, or abuse detection
- Experience working with scarce, delayed, or noisy labels
- Experience with AutoML or other ML workflow automation approaches
- Care about societal impacts of AI and desire to make powerful systems safer
