Data Scientist - Fraud Strategy Associate

Data Scientist - Fraud Strategy Associate

Goldman Sachs' Asset & Wealth Management (AWM) offers a unique opportunity within a premier global financial institution. We support a diverse clientele, including mutual funds, hedge funds, pension plans, sovereign wealth funds, insurance companies, endowments, foundations, third-party wealth firms, and ultra-high-net-worth individuals. With over $3 trillion in assets under supervision, AWM provides innovative investment and advisory services across traditional and alternative investments, prioritizing long-term performance and client success.

Marcus by Goldman Sachs

Marcus, the digital consumer banking arm of Goldman Sachs, offers high-yield savings accounts and Certificates of Deposit directly to consumers. Leveraging 150+ years of Goldman Sachs expertise with intuitive digital experiences, Marcus focuses on value, transparency, and simplicity for its millions of customers, operating as the largest pure online bank without physical branches.

Responsibilities:

• Analyze extensive data using advanced statistical techniques to identify new fraud patterns and conduct in-depth reviews.
• Design and develop data-driven fraud strategies and capabilities for consumer money movement products to mitigate fraud losses.
• Utilize supervised and unsupervised machine learning to accurately detect high-risk activities on customer accounts.
• Build new data features and products to enhance statistical fraud models.
• Identify data signals to effectively differentiate between fraudulent and non-fraudulent activities.
• Explore and integrate new data sources for robust fraud control development.
• Generate trend reports and analyses using Python, PySpark, SQL, Snowflake, Databricks, and Excel.
• Synthesize current portfolio risk and trend data to inform recommendations.
• Investigate and leverage cloud-based data science technologies to advance fraud controls.
• Measure and monitor the impact of risk controls on customers and develop strategies for a positive customer experience.
• Collaborate with technology and capability partners to implement innovative data-driven solutions.

Basic Qualifications

• Bachelor’s degree in Mathematics, Statistics, Economics, Finance, Engineering, or a related field.
• Proven experience with large datasets using Big Data tools and platforms (e.g., Python, PySpark, Snowflake, Databricks, SQL).
• Ability to derive key insights and signals from complex structured and unstructured data.
• Strong working knowledge of statistical techniques including regression, clustering, neural networks, and ensemble methods.
• 2+ years of experience in fraud risk management, preferably with banking products (savings, checking, CDs, credit cards).
• Creativity in developing solutions beyond standard tools and comfort working independently.
• Demonstrated thought leadership, creative thinking, and project management skills.

Preferred Qualifications

• Master’s degree in Mathematics, Statistics, Economics, Finance, Engineering, or a related field.
• Experience building quantitative, data-driven statistical strategies for consumer checking and savings businesses.
• Familiarity with large-scale graph processing, including graph clustering and link prediction algorithms.
• Expertise in advanced machine learning techniques (ensemble methods, reinforcement learning, deep neural networks).
• Knowledge of fraud risk vendors and technologies in consumer finance or digital services.
• Experience with consumer banking authentication tools and methodologies.
• Experience in reporting and data visualization tools for trend analysis.

Key skills/competency

• Data Science
• Fraud Strategy
• Machine Learning
• Python
• PySpark
• SQL
• Snowflake
• Databricks
• Statistical Analysis
• Risk Management