Data scientist ML engineer - Freelance

Context

Decathlon Digital aims to become the best digital sports platform and open ecosystem worldwide. The company focuses on enabling customers to experience Decathlon through numerous local sport-centric experiences by connecting many third-party actors and services securely and efficiently. Decathlon Digital includes over 5000 technical profiles such as software engineers, product managers, data, cloud, and cybersecurity experts located in Paris, Lille, Nantes, Lyon, and Amsterdam.

The Inventory Optimization team, part of the Data Supply Chain, is responsible for designing, developing, and maintaining innovative analytical and Artificial Intelligence solutions related to supply chain use cases like multi-echelon demand forecasting, target stock estimation, and shortage management. The team works in an agile environment alongside Data Scientists, Data Analysts, Analytics Engineers, ML Engineers, Product and Engineering contacts, and business users.

Missions

• Analyze, develop, and deploy Inventory Optimization solutions at scale.

• Develop and deploy production-grade ML models for demand forecasting across BTB and BTC channels at various geographical granularities (stores, warehouses).

• Build robust models integrating internal variables (price, assortment) and external signals (weather, calendar data, market trends) for global replenishment.

• Conduct in-depth diagnostic analyses of production forecasts and user-driven overrides to identify patterns, refine model accuracy, accelerate deployment, and drive automation.

• Manage the end-to-end lifecycle of ML solutions from R&D to CI/CD deployment, ensuring forecasting engine stability and efficiency.

• Translate complex data insights into actionable recommendations for Supply Chain and Business stakeholders using compelling visualization and storytelling.

• Participate actively in daily technical initiatives, contribute to codebases, remove blockers, and maintain high team velocity.

• Engage in code and model reviews to promote best practices in ML engineering, reproducibility, and documentation.

• Collaborate with team members to foster a culture of technical excellence and knowledge sharing.

Tools & Environment

• Execution Engine: Databricks, AWS, GCP

• Programming Languages & Libraries: Python, Spark, Scikit-Learn, TensorFlow / PyTorch, PySpark

• CI/CD: GitHub Actions

• Serving: Docker, Protobuf, gRPC

• Model Registry & Tracking: MLFlow

• Orchestration: Airflow

• Documentation & Code Management: Git, Confluence

• Data Visualization: Tableau

Working Conditions

• Location: Paris