Senior ML Engineer

Xebia is a global AI-first, digital transformation, and engineering partner. With over 25 years of experience and a team of 5,000 professionals across 16 countries, we help organizations design and build scalable products, platforms, and data-driven solutions.

You will:

take ownership of an existing ML solution and codebase following knowledge transfer from the Lead AI/ML Engineer,

develop and productionize machine-learning models for new-product first-order recommendations,

build and maintain training and inference pipelines using Azure Machine Learning,

implement scheduled model retraining and integrate models with the existing model registry and validation gate,

implement data quality and completeness checks, including validation of product attributes required by the models,

develop recommendation logic accounting for service levels, minimum order quantities, time until the next delivery and category-based fallback rules,

implement and validate the transition of products from the new-item model to the main forecasting model as sufficient sales history becomes available,

extend daily recommendations to eligible catalogue products based on agreed criteria and backtesting results,

support the phased rollout of catalogue products and implement monitoring for individual product segments,

maintain MLOps capabilities, including model versioning, validation, automated retraining, monitoring, data-drift detection and alerting,

integrate ML outputs with the existing data warehouse and Power BI reporting,

ensure production runs are reproducible and that model versions retain their training windows and feature contracts,

participate in testing, dry runs, parallel runs, documentation and operational handover,

collaborate with the Lead AI/ML Engineer, BI team and business stakeholders to ensure a reliable production rollout.

Your profile:

hands-on experience developing and productionizing machine-learning solutions,

practical experience with Azure Machine Learning, including training and inference pipelines,

experience implementing MLOps practices, including model registries, model versioning, automated retraining, validation gates, monitoring and rollback,

experience building and maintaining scheduled batch ML pipelines,

ability to take ownership of an existing ML codebase and continue development following an initial knowledge-transfer phase,

experience integrating ML pipelines with data warehouses,

experience implementing data validation and input-quality checks,

experience with model monitoring and data-drift detection,

experience implementing automated testing and CI/CD for ML workloads,

familiarity with Git-based repositories and deployment workflows,

understanding of model evaluation, backtesting, champion/challenger validation and production acceptance criteria,

ability to document ML systems and operational procedures,

ability to collaborate effectively with technical teams and business stakeholders,

ability to work independently on implementation after an initial knowledge-transfer period,

availability to participate in an approximately 15-week MVP, with an expected commitment of around 45 engineer-days.

Nice to have:

experience with demand forecasting, inventory optimization or purchasing and replenishment systems,

experience with LightGBM,

experience working with Azure SQL-based data warehouses,

familiarity with Azure managed identities, Key Vault and Azure Monitor,

familiarity with Power BI integrations and ML monitoring and reporting use cases,

experience with Azure DevOps or GitHub Actions and Azure deployment workflows,

experience working with service-level-based inventory or order-quantity models,

familiarity with model calibration and monitoring forecast accuracy and bias,

experience developing ML solutions for new-product forecasting with limited historical data,

familiarity with retail, wholesale, supply chain or catalogue-based business environments.