5 hours ago
Base Salary
$229k - $315k/yr
Responsibilities
- Investigate fraud patterns and model errors to identify new signals, improve detection, and expand coverage.
- Develop training datasets and predictive features while addressing incomplete labels, class imbalance, data leakage, and changing fraud behavior.
- Design, train, tune, and evaluate models including gradient-boosted trees and neural networks.
- Design experiments comparing model performance across time periods and customer segments using detection and false-positive metrics.
- Build reproducible data and training pipelines for efficient experimentation and iteration.
- Deploy models with Engineering and ML Infrastructure partners while balancing detection quality, latency, cost, and reliability.
- Lead ML projects independently from initial experimentation through deployment, model release, and ongoing improvement.
- Evaluate production model impact using real-world customer outcomes and explore LLMs and Generative AI for fraud detection, prevention, and investigation.
Requirements
- 7+ years of professional experience in machine learning, applied science, or software engineering for ML.
- Hands-on experience designing, training, tuning, deploying, and evaluating machine learning models in production.
- Strong machine learning and statistical fundamentals, including feature engineering, experiment design, model evaluation, and diagnosing underperforming models.
- Understanding of traditional and modern ML methods, including gradient-boosted trees and neural networks.
- Experience constructing training datasets and addressing label quality, data leakage, class imbalance, and generalization across time periods or populations.
- Strong Python skills and SQL proficiency for training and evaluation data.
- Hands-on experience with ML frameworks such as PyTorch, scikit-learn, XGBoost, or equivalents.
- Experience independently leading open-ended ML projects through deployment and coordinating requirements and releases with Data Science, Product, and Engineering.
- Preferred experience in fraud or risk modeling, including delayed feedback and balancing fraud detection with legitimate-user friction.
- Preferred experience developing models that generalize across customers and using graph-based systems to extract predictive signals.
- Preferred experience applying learned representations, transformers, or foundation models to production ML use cases.
Benefits
- Comprehensive medical, dental, vision, and 401(k) benefits.
- Additional compensation may include equity and/or commission, depending on the position offered.
- Pay and benefits are based on factors including scope, responsibilities, experience, skills, and location and may change under applicable plans.
Categories
About Plaid
Plaid builds APIs and SDKs that let developers and financial institutions connect consumer bank and investment accounts, verify ownership, retrieve transactions, assess income/liabilities, fight fraud, and move money (e.g., ACH/transfer). It sells usage-based access to products like Link, Auth, Transactions, Identity, Income, Liabilities, and Payments. Founded in 2013 and headquartered in San Francisco, Plaid’s network spans 12,000+ financial institutions across the US, Canada, the UK, and Europe, with customers including Venmo, SoFi, and Betterment.