Were looking for an Applied Data Scientist to join one of our product squads. Youll design, build, and deploy data-driven solutions that combine machine learning, statistical methods, and SQL/rules-based decision logic to power autonomous supply chain intelligence platform. Youll work closely with data science, engineering, product, and supply chain experts and own solutions end-to-end-from problem definition to production monitoring and iteration.
Responsibilities:
Deliver data science solutions end-to-end within a product squad: problem framing → data prep/labeling → modeling → deployment support → monitoring → iteration
Build, train, and improve ML models for supply chain use cases (e.g., inventory risk prediction, demand anomalies, root-cause analysis)
Define success metrics and evaluation plans with support from senior DS/PM; run error analysis and document learnings
Work with stakeholders to create and maintain ground truth (label definitions, labeling workflows, QA checks, feedback loops)
Implement hybrid decision logic by combining ML outputs with statistical methods and SQL/rules-based logic for robustness and explainability
Analyze large, multi-source operational datasets to identify trends, anomalies, and drivers of performance
Collaborate with software engineers to productionize solutions (batch and/or real-time), including testing, logging, and basic monitoring
Monitor deployed models/rules, investigate performance issues (data quality, drift, edge cases), and iterate based on outcomes
Contribute to team practices: reproducible notebooks/code, documentation, and experiment tracking
Requirements: MSc in Computer Science, Data Science, Mathematics, Statistics, Engineering, (or equivalent practical experience)
3+ years of experience in applied data science / ML in a product environment (or equivalent practical experience)
Strong Python skills and experience with common DS libraries (pandas, NumPy, scikit-learn); familiarity with PyTorch/TensorFlow is a plus
Solid SQL skills (joins, aggregations, window functions) and comfort working with production data in a warehouse/lake
Experience building predictive or anomaly detection models and performing rigorous evaluation (baselines, cross-validation where relevant, error analysis)
Ability to translate business questions into measurable metrics and a clear analytical plan (with guidance when needed)
Experience working with messy real-world data: data validation, debugging pipelines, and collaborating on labeling/ground truth
Familiarity with taking models to production: packaging/hand-off to engineers, versioning, and understanding monitoring/drift concepts
Strong communication and collaboration skills with engineering, product, and domain experts; comfortable receiving feedback and iterating fast
Nice to Have (Advantages)
Experience designing or deploying agentic workflows, AI agents, or multi-step decision systems
Cloud + Docker + production engineering practices (CI/CD, testing, monitoring)
Experience publishing academic or applied research (peer-reviewed papers, conference publications, technical whitepapers, or open research work)
This position is open to all candidates.