Staff Machine Learning Engineer

PayPal • San Jose, CA

1d • $196,833 - $333,500 • Remote

About The Position

PayPal, Inc. seeks a Staff Machine Learning Engineer to lead the design, development, and implementation of advanced machine learning models and algorithms for complex problems. The role works closely with data scientists, software engineers, product teams, and other partners to build scalable ML pipelines, ensure data quality, and deploy models into production to improve services and customer experiences.

The position manages the FSI Data Integration Project with Risk Models, coordinating with infrastructure and online engineering teams to expose relevant features at system checkpoints and validating their online availability and quality. It partners with Global Analytics, Data Science, and model developers to enumerate and compare features, measure their impact, and support their adoption across fraud and risk models, including account takeover, collusion, seller risk, and credit risk.

The role also develops risk models based on linking features, evaluates whether an internal graph-based tool can support FSI objectives, investigates embedding features, and builds and tests proof-of-concept and automated solutions. Additional work includes transforming FSI data into graph infrastructure, applying graph data science insights to linking automation, and automating detection of new risky features. Telecommuting is permitted, and work may be performed from anywhere in the United States. The listed schedule is 40 hours per week, Monday through Friday, 9:00 a.m. to 5:00 p.m.

Requirements

• Master’s degree or foreign equivalent in Computer Science, Engineering (any field), or a closely related field, plus four years of experience in the job offered or a related occupation; alternatively, a bachelor’s degree or foreign equivalent in one of these fields plus six years of such experience.
• Four years of experience in financial fraud detection, including developing machine learning models to identify fraudulent activity.
• Four years of experience translating business stakeholder requirements into machine learning problem formulations, including defining KPIs, evaluation metrics, and success criteria.
• Four years of experience designing and developing end-to-end machine learning solutions, including data collection, feature engineering, preprocessing, feature selection, research, model architecture, training, optimization, evaluation, model encoding, production deployment, and maintenance.
• Four years of experience automating end-to-end machine learning solutions, including automated data collection, model training, model encoding, and production deployment.
• Four years of experience with Python-based data science libraries, including scikit-learn, pandas, NumPy, and Keras, for preprocessing, training, and evaluating machine learning models.
• Four years of experience with large-scale tabular datasets in cloud-based data warehouse environments for feature engineering and analytics.
• Four years of experience encoding and optimizing machine learning models for production deployment using serialization frameworks such as ONNX for cross-platform inference.
• One year of experience in embedding representation methods for tabular data.
• Four years of experience collaborating with cross-functional business strategy, engineering, data infrastructure, and data science teams to align model outputs with business objectives and integrate models into production decision systems.
• Must be legally authorized to work in the United States without sponsorship.

Responsibilities

• Manage the FSI Data Integration Project with Risk Models.
• Coordinate with infrastructure and online engineering teams to expose relevant features at system checkpoints and verify their existence and quality as online variables.
• Work with Global Analytics and Data Science teams to enumerate features and compare them to identify measurable uplift from FSI-engineered features.
• Scale feature integration across models for fraud domains such as account takeover, collusion, seller risk, and credit risk.
• Develop plans with model developers so they can use FSI features, and assist them in resolving issues.
• Work with the team that developed LEGO, an internal graph-based tool for generating online variables, to assess its suitability for FSI objectives.
• Investigate embedding features as a potential solution for representing FSI intelligence in a small number of features.
• Develop a proof of concept and compare its results with the current solution.
• Develop the complete automated tool and conduct experiments to verify its operation and monitor its performance.
• Transform FSI data into graph infrastructure.
• Apply graph data science insights to FSI linking automation.
• Automate the detection of new risky features.
• Design, develop, and implement machine learning models and algorithms; build scalable ML pipelines; ensure data quality; and deploy models into production.

Benefits

• Generous paid time off.
• Healthcare coverage for the employee and family.
• Resources to support financial security and mental health.
• Annual performance bonus, equity, or other incentive compensation, as applicable.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior