As part of the AI Operations Integration team, we're passionate about pushing the boundaries of AI and transforming how operations teams work. We are looking for an entrepreneurial, experienced, creative, and AI-Native Data Scientist I to join our team. As a Data Scientist I on the AI Operations Integration team, you'll have the opportunity to work on exciting, ambiguous problems that combine Large Language Models (LLMs), Generative AI, and predictive analytics to create intelligent, data-driven operational solutions that fundamentally change how work gets done across Amazon's global operations footprint.
You will be responsible for leading the development and delivery of core data science capabilities that power AI-enabled operations. You will have significant influence on our overall strategy by defining analytical approaches, driving solution architecture, and spearheading the data science best practices that enable a high-quality, scalable AI ecosystem.
In this role, you'll collaborate with a diverse team of software engineers, AI/ML specialists, operations experts, and technical program managers to develop novel solutions that advance the state of the art in AI-enabled operations. You'll leverage Amazon's vast data resources and computing infrastructure to accelerate development and drive innovation. Your contributions will help define our overall data science strategy, from data enrichment and model optimization to system architecture and best practices, creating a virtuous cycle of AI-enablement that continuously improves operational excellence.
Key job responsibilities
- Assess and select ideal solution approaches from a wide range of data science methodologies, including machine learning, statistical modeling, NLP, and LLM-based techniques, to solve complex, ambiguous operational problems with significant business impact.
- Apply deep expertise to problems involving complex interactions among software systems, data pipelines, and operational processes; design solutions that accurately model these interactions and are extensible, actionable, and easy for others to contribute to.
- Own and deliver end-to-end data science solutions for the business with minimal assistance, building a track record of successful launches that drive measurable operational improvements across Amazon's global footprint.
- Work closely with operations business teams to deeply understand their challenges, translate ambiguous needs into well-defined problem statements, and ensure data science solutions are grounded in real operational context.
- Take the lead on large, cross-functional data science initiatives; drive solutions and influence change across multiple teams connected by shared systems and processes; build consensus among discordant views and align stakeholders on the right path forward.
- Make sound scientific and technical trade-offs to meet both short-term operational needs and long-term technology sustainability goals; advocate for the right measurements, sensors, and metadata to ensure solutions are built on reliable signal.
- Stay current on data science developments and emerging research; raise awareness of new and well-established techniques across the team; lead knowledge-sharing sessions and mentor data scientists at all levels to help develop the best.
- Drive data science best practices, set standards, and proactively lead initiatives to improve operational excellence; identify blind spots in current metrics, challenge assumptions, and restructure data sources to better reflect operational reality.
- Partner with engineering and AI/ML teams to integrate data science solutions into existing operational systems; contribute to strategic planning (OP1/QBR/MBR) and advise senior leadership on AI investment priorities and data science strategy.
A day in the life You start your morning with a profitability puzzle. Thousands of low-price products are losing money, and no single team can explain why. The buying, placement, and fulfillment systems each say they did the right thing, but the customer's order still ships in three boxes from three warehouses. You trace decisions across systems, find that a parameter was quietly misconfigured weeks ago, and write up the evidence chain.
Later you dig into a natural experiment, a recent policy change gave some products broader warehouse coverage. You run a causal analysis to test whether that actually improved shipment consolidation, check the assumptions, and document what you find with confidence intervals and boundary conditions. Not everything is a clean win: the effect is real for products customers buy together, but disappears for standalone items.
A couple times a week, you join a cross-team working session where scientists, engineers, and data teams collaborate on end-to-end investigations. You're connecting the dots across systems that don't normally talk to each other tracing a product from purchase order to customer doorstep and pinpointing where value leaks. Some cases have obvious fixes. The more interesting ones are where every system worked as designed but the outcome is still bad.
On other days you might build a counterfactual simulation to test whether a different optimization approach would change the economics, design an A/B test to validate it, or present findings to leadership walking them through what you know, what you don't, and what level of confidence each finding carries.
The thread that connects it all: you're turning complex cross-system problems into structured evidence that people can act on. Some of that is causal inference, some is building AI-assisted investigation tools (and figuring out where AI helps vs. where it confidently gives you the wrong answer), and some is just good old-fashioned detective work across messy operational data.
About the team We're part of a broader organization transforming how global operations teams work through AI. Within that mission, our team focuses on the hardest diagnostic problems: when automated supply chain systems produce bad outcomes and no single team can explain why. We build decision intelligence platforms that traces decisions across automated systems and uses causal engines and AI to find root causes. You'll work alongside scientists, SDEs, and ML engineers, and collaborate regularly with cross-functional partner SMEs. The team is new and you'd help shape it from the ground up.
Basic Qualifications
- Experience with data scripting languages (e.g., SQL, Python, R, or equivalent) or statistical/mathematical software (e.g., R, SAS, Matlab, or equivalent)
- Experience as a data/research scientist, statistician or quantitative analyst in an internet-based company with complex and big data sources
Preferred Qualifications
- Knowledge of statistical packages and business intelligence tools such as SPSS, SAS, S-PLUS, or R
- Experience with clustered data processing (e.g., Hadoop, Spark, Map-reduce, and Hive)
Amazon is an equal opportunities employer. We believe passionately that employing a diverse workforce is central to our success. We make recruiting decisions based on your experience and skills. We value your passion to discover, invent, simplify and build. Protecting your privacy and the security of your data is a longstanding top priority for Amazon. Please consult our Privacy Notice (https://www.amazon.jobs/en/privacy_page) to know more about how we collect, use and transfer the personal data of our candidates.
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.