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 Amazons 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 Amazons 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 Amazons 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 Amazons compliance framework.

RCSs mission is to apply specialist industry insights and Amazons unique approach to scale through technology delighting customers through tailored compliance solutions. Our work spans multiple critical areas including:
Enabling Amazons 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

- 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
- Masters 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

- 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 for more information. If the country/region youre applying in isnt listed please contact your Recruiting Partner.