(IND) STAFF, DATA SCIENTIST
Walmart
Position Summary...
As a Staff Data Scientist on Scintilla, you will lead the end-to-end delivery of recommendation and decisioning systems that convert Walmart’s first-party signals into actionable, measurable outcomes for suppliers and merchants. You will drive multiple complex projects simultaneously by shaping the problem, aligning stakeholders, building robust models, and ensuring production excellence through strong MLE and MLOps practices.This is a senior individual contributor role that requires technical leadership, strong product intuition, and the ability to move from research to production with speed and reliability. Scintilla’s mission is to turn granular data into insights that help improve decisions across the retail value chain, from customer experience to operational excellence.
What you'll do...
About the Team (Scintilla)
Scintilla is the flagship platform from Walmart Data Ventures, designed to transform Walmart’s first-party customer and operational data into supplier and merchant intelligence. Over time, the ecosystem has expanded across modules such as Shopper Behavior (basket visibility), Channel Performance, Customer Perception, and Digital Landscapes (pre-purchase behavior on Walmart.com and the Walmart app).
Digital Landscapes focuses on pre-purchase behavior, helping reveal how customers search for, consider, and select products, including signals like source of traffic (external pages vs Walmart search).
Shopper Behavior provides first-party insights into brand and category behavior and positions basket data as a shared source of truth used to inform decisions on products and promotions.
Scintilla also extends into activation and execution. For example, Insights Activation is positioned as an automated way to extract insights and generate activation opportunities, including audience creation and optimization workflows.
And Scintilla In-Store brings the ecosystem into stores, unifying real-time data, metrics, and tasks to help reduce out-of-stocks and streamline execution, with an emphasis on future AI-driven prioritization.
What You’ll Do
1) Own recommendation systems end-to-end (primary)
Lead the development of recommendation, ranking, and decisioning systems from problem definition through production rollout.
Build systems across the full funnel, for example: candidate generation, retrieval, ranking, re-ranking, and explainability strategies.
Design recommendation strategies for multiple surfaces and use cases, such as next-best-action recommendations, product association, shopper journey interventions, or supplier-facing recommendations where applicable.
2) Drive problem formulation, success metrics, and experimentation
Translate ambiguous business problems into precise ML problem statements with measurable success criteria.
Define offline and online evaluation methodologies (offline metrics, A/B testing, guardrails, and iteration loops).
Partner with Product and Engineering to align on trade-offs (latency, cost, quality, interpretability) and set launch criteria.
3) Deliver production-grade ML systems with strong MLE/MLOps
Build robust data pipelines and feature engineering workflows (batch and near real-time), ensuring high data quality and governance.
Own model productionization, including CI/CD, model registry, monitoring, drift detection, performance analytics, and incident response playbooks.
Establish quality standards across testing, evaluation, and reliability so models are safe to scale.
4) Act as a science generalist when needed
Apply a broad toolkit beyond recommenders when the business problem demands it: causal inference, forecasting, segmentation, optimization, anomaly detection, and measurement.
Connect insights to action by designing decision frameworks rather than one-off analyses.
5) Use GenAI and Agentic AI as a secondary skillset
Apply GenAI selectively to improve user workflows, for example: natural-language interfaces, insight summarization, recommendation explanation, retrieval-augmented experiences, and agentic workflows.
Build evaluation and safety guardrails for LLM-based systems (hallucination checks, grounding, citation strategy, and feedback loops).
6) Lead through influence
Collaborate cross-functionally across Product, Engineering, UX, and business stakeholders.
Mentor junior data scientists and raise team standards in modeling, experimentation, and production readiness.
Drive best practices for documentation, design reviews, and operational excellence.
Core Technical Competencies (Staff Level Expectations)
Tech Problem Formulation
Frames and decomposes complex recommendation problems into solvable ML tasks with clear constraints, assumptions, and measurable outcomes.
Business and Product Context
Connects supplier and merchant workflows to the data products and recommendations we build, ensuring outputs are interpretable and actionable.
Data Source Identification and Data Quality
Identifies the right first-party signals (behavioral, transactional, digital journey, and operational) and establishes strong quality and governance practices.
Analytical Modeling (Recommendations first)
Deep expertise in recommendation systems: collaborative filtering, matrix factorization, embeddings, graph-based methods, sequential recommenders, learning-to-rank, bandits where relevant.
Strong generalist foundation: statistics, experimentation, causal inference, and optimization.
Model Assessment and Validation
Designs rigorous offline evaluation and online experimentation strategies with robust guardrails.
Understands failure modes, slice analysis, and bias and fairness considerations.
Model Deployment and Scaling (MLE/MLOps)
Production model serving patterns, monitoring, drift detection, reliability engineering, and cost-performance optimization.
Code Development and Testing
Writes production-quality code, establishes test coverage, code reviews, and clean interfaces in partnership with engineering.
Data Visualization and Storytelling
Creates clear, decision-driving narratives and visuals for technical and non-technical audiences.
Data Strategy
Builds reusable datasets, feature definitions, and evaluation harnesses that compound impact across multiple products and teams.
What You Will Bring
Bachelor’s degree in Computer Science, Statistics, Mathematics, Economics, Engineering, or related field.
7+ years of experience in applied data science or machine learning roles, with meaningful time spent building and deploying models into production.
Demonstrated experience owning recommendation systems or ranking systems end-to-end, including evaluation and iteration in production.
Strong hands-on expertise in Python and SQL, and experience working with large-scale data and distributed compute (Spark or similar).
Proven ability to partner with engineering teams to productionize models, including MLOps and operational ownership.
Preferred Qualifications
9 to 10+ years of experience building large-scale personalization, recommendation, search, or ranking systems.
Experience with graph-based recommenders, embeddings at scale, ANN retrieval, and real-time personalization constraints.
Strong MLOps depth: CI/CD for ML, model registry, monitoring, feature stores, model governance, and incident management.
Experience with GenAI systems (RAG, tool-using agents) and strong evaluation discipline for grounded outputs.
Prior experience delivering data products or ML systems in retail, e-commerce, ads, marketplace, or supply chain contexts.
About Walmart Global Tech
Imagine working in an environment where one line of code can make life easier for hundreds of millions of people. That’s what we do at Walmart Global Tech. We’re a team of software engineers, data scientists, cybersecurity expert's and service professionals within the world’s leading retailer who make an epic impact and are at the forefront of the next retail disruption. People are why we innovate, and people power our innovations. We are people-led and tech-empowered.
We train our team in the skillsets of the future and bring in experts like you to help us grow. We have roles for those chasing their first opportunity as well as those looking for the opportunity that will define their career. Here, you can kickstart a great career in tech, gain new skills and experience for virtually every industry, or leverage your expertise to innovate at scale, impact millions and reimagine the future of retail.
Ways of working
Walmart’s culture sets us apart, and we know being together helps us innovate, learn and grow great careers. This role is based in our Bangalore office for daily work, with the flexibility for associates to manage their personal lives
Benefits
Beyond our great compensation package, you can receive incentive awards for your performance. Other great perks include a host of best-in-class benefits maternity and parental leave, PTO, health benefits, and much more.
Equal Opportunity Employer:
Walmart, Inc. is an Equal Opportunity Employer – By Choice. We believe we are best equipped to help our associates, customers and the communities we serve live better when we really know them. That means understanding, respecting and valuing diversity- unique styles, experiences, identities, ideas and opinions while being inclusive of all people.
Minimum Qualifications...
Outlined below are the required minimum qualifications for this position. If none are listed, there are no minimum qualifications.
Minimum Qualifications:Option 1: Bachelors degree in Statistics, Economics, Analytics, Mathematics, Computer Science, Information Technology or related field and 4 years' experience in an analytics related field. Option 2: Masters degree in Statistics, Economics, Analytics, Mathematics, Computer Science, Information Technology or related field and 2 years' experience in an analytics related field. Option 3: 6 years' experience in an analytics or related field.Preferred Qualifications...
Outlined below are the optional preferred qualifications for this position. If none are listed, there are no preferred qualifications.