stripe
Machine Learning Engineer, Link
About this role
Stripe is seeking a Machine Learning Engineer to build and operate fraud detection and risk decisioning models for Link, their digital wallet platform. You'll own the full ML lifecycle—from analyzing fraud patterns to deploying and monitoring production models—while collaborating across engineering, product, and data science teams to protect Link users and maximize legitimate payment authorization rates.
What you'll do
- Build, train, evaluate, and deploy ML models that detect fraud and abuse across Link's payment network
- Investigate emerging fraud threats using large-scale datasets and develop hypotheses to improve payment performance
- Design data pipelines, feature engineering, and monitoring systems to support reliable production models
- Develop pragmatic ML solutions optimized for real-time risk decisioning, including tree-based models
- Build and improve risk decisioning systems that integrate with Stripe's broader payments infrastructure
- Own end-to-end ambiguous problems from analysis through implementation, launch, and iteration
What they're looking for
- Machine learning model development and deployment in production
- Python programming and data engineering (SQL, Spark, XGBoost)
- Fraud detection and adversarial ML domain knowledge
- Real-time, low-latency systems design and optimization
- Data analysis, statistics, and A/B testing
- Feature engineering and data pipeline design
- Production ML systems (monitoring, evaluation, iteration)
- Cross-functional collaboration and technical communication
Opens the official application on the employer’s site. No login required.
stripe
Stripe builds payment infrastructure and financial services platforms, offering APIs and tools that enable developers and businesses to process transactions, detect fraud, verify identity, and manage security at scale. The company is hiring Backend Engineers, Full Stack Engineers, ML Engineers, AI Engineers, and Security Engineers to develop core platform systems, payment intelligence, customer support infrastructure, and security data platforms.
- Website
- stripe.com
Likely interview questions
- Tell us about a fraud detection or risk modeling project you shipped in production—what were the key challenges and how did you measure success?
- How do you approach selecting between different modeling approaches (e.g., tree-based vs. neural networks) when building real-time decisioning systems?