Data Scientist II, Risk and Compliance Solutions (RCS)

Can you predict which compliance risks turn into breaches before they do? What would you build with control testing results, alert data, and investigation outcomes from across Amazon's regulated entities in one place? How would you define the target variable when a breach means different things to Compliance, Assurance, and the regulator?

We are seeking a Data Scientist to join the Data and Analytics team within Risk and Compliance Solutions (RCS). Our team builds the analytics and modeling layer that RCS runs on: the datasets, models, and intelligence products that tell Compliance Officers and our first line of defense (1LoD) partners where risk concentrates and which controls hold. You build the predictive side of that layer. You score which open risks and control weaknesses convert to breaches, quantify control effectiveness across regions, and put the answer in front of the people who act on it.

Financial crime compliance is the first domain where we apply this, and the pattern extends across the RCS portfolio from there. Threats change faster than control sets adapt, and the intelligence that explains a control failure in one entity rarely reaches the teams running the same controls elsewhere. You close that gap with models, and you carry the approach into the next domain once it proves out.

You own problems end to end: framing the question with Compliance and 1LoD partners, building and calibrating models on control, alert, and investigation data, and moving the output into production reporting. You define what counts as a breach before you model it, which means anchoring event dates to control testing cycles, designing out temporal leakage, and mapping the target variable to the taxonomy Compliance and Assurance already recognize. You work with technologies including Amazon Redshift, Amazon SageMaker, AWS Lambda, and Amazon QuickSight, and you partner with data engineers, business intelligence engineers, and software development engineers across RCS.

The successful candidate reasons about causality rather than correlation, defends a model to an audience that answers to regulators, and stays comfortable in problems where defining the target variable is the hardest part of the work.

Key job responsibilities
• Build, deploy, and monitor predictive models that identify which risks and control weaknesses convert to breaches, and calibrate them so a score carries a defensible meaning
• Define target variables and labels with Compliance and 1LoD partners so model output maps to the risk taxonomy already in use
• Combine control testing results, alert and escalation data, investigation outcomes, and gross merchandise sales at risk into analysis-ready datasets
• Quantify control effectiveness across regions and separate genuine control failure from shifts in underlying exposure
• Design backtesting against historical events and select evaluation metrics suited to rare outcomes, including precision-recall and calibration measures
• Design and evaluate the analytics behind a joint testing environment where first line and second line teams test monitoring rule and risk model changes before release
• Turn model output into intelligence that reaches Compliance Officers and 1LoD teams across regions through forums, briefings, and alerting
• Apply generative artificial intelligence (GenAI) where it removes manual effort in control testing, requirement translation, and reporting
• Present model results, assumptions, and limitations to Compliance and business audiences, and document methodology to a standard that holds up under internal audit and external examination
• Extend proven approaches from the first compliance domain to the next, so each model becomes a reusable capability rather than a one-time analysis
• Learn Amazon's broad range of data resources and know how, when, and what to use

About the team
Risk and Compliance Solutions (RCS) serves as a crucial safeguard for Amazon's buyers, brands, and selling partners while ensuring business teams meet their regulatory obligations. Operating under the Three Lines of Defense (3LoD) model, RCS functions as the second line of defense, providing specialized oversight, monitoring, and expertise while setting policies and procedures that guide Amazon's compliance framework.

RCS's mission is to apply specialist industry insights and Amazon's unique approach to scale through technology, delighting customers through tailored compliance solutions. Our work spans multiple critical areas, including:
• Enabling Amazon's financial services expansion through streamlined license acquisition
• Managing existing business risks across the organization
• Delivering comprehensive governance, risk, and assurance services
• Developing innovative solutions like Compliance as a Service (CaaS)
• Transforming compliance operations into a proactive, intelligence-led framework

Basic Qualifications

- 2+ years of data scientist experience
- 3+ years of data querying languages (e.g. SQL), scripting languages (e.g. Python) or statistical/mathematical software (e.g. R, SAS, Matlab, etc.) experience
- 3+ years of machine learning/statistical modeling data analysis tools and techniques, and parameters that affect their performance experience
- 1+ years of guiding and coaching a group of researchers experience
- 1+ years of working with or evaluating AI systems experience
- 1+ years of creating or contributing to mathematical textbooks, research papers, or educational content experience
- Master's degree in Science, Technology, Engineering, or Mathematics (STEM), or experience working in Science, Technology, Engineering, or Mathematics (STEM)
- Experience applying theoretical models in an applied environment

Preferred Qualifications

- Ph.D. in Science, Technology, Engineering, or Mathematics (STEM)
- Knowledge of machine learning concepts and their application to reasoning and problem-solving
- Experience in Python, Perl, or another scripting language
- Experience in a ML or data scientist role with a large technology company
- Experience in defining and creating benchmarks for assessing GenAI model performance
- Experience working on multi-team, cross-disciplinary projects
- Experience applying quantitative analysis to solve business problems and making data-driven business decisions
- Experience effectively communicating complex concepts through written and verbal communication

Our inclusive culture empowers Amazonians to deliver the best results for our customers. If you have a disability and need a workplace accommodation or adjustment during the application and hiring process, including support for the interview or onboarding process, please visit https://amazon.jobs/content/en/how-we-hire/accommodations for more information. If the country/region you’re applying in isn’t listed, please contact your Recruiting Partner.