Senior Data Scientist - at Kamino Retail - Full-remote (France)

📋 Description

• • Own the birth of Kamino Retail’s data-science brain. You will design, train and deploy the very first deep-learning algorithms that turn billions of daily logs into predictive power, giving retailers the smartest ad-targeting and customer-engagement engine on the market.
• • Translate complex retail-media business questions—such as “Which shopper is most likely to convert after seeing a sponsored placement?”—into crisp machine-learning problems, then select the right statistical or neural approach to solve them.
• • Architect end-to-end ML pipelines that run 24/7 on our Kubernetes cluster: data ingestion from ClickHouse, feature engineering in Python, model training with GPU acceleration, automated validation, canary release and real-time inference.
• • Implement rigorous MLOps practices from day one: versioned datasets, experiment tracking, model registry, CI/CD for training jobs, and live performance dashboards so every stakeholder sees how models behave in production.
• • Collaborate hand-in-hand with our dedicated data engineer to co-design scalable data marts, feature stores and monitoring tools; you will jointly decide storage formats, partitioning strategies and compute budgets to keep latency low and costs lean.
• • Champion a product-focused mindset: iterate quickly with A/B tests, quantify uplift in revenue or ROAS, and ship only what demonstrably improves the customer experience—no “research for research’s sake.”
• • Influence the roadmap by translating model results into clear business narratives for retail clients, product managers and executives; your insights will directly steer which new capabilities we build next.
• • Mentor future hires and set the technical bar for code quality, peer review and documentation, ensuring our growing data team remains cohesive and world-class.
• • Stay ahead of the curve by evaluating emerging techniques—transformers for sequence prediction, graph neural networks for shopper graphs, reinforcement learning for dynamic pricing—and run internal spikes that keep Kamino Retail at the cutting edge.
• • Enjoy true ownership: from whiteboard math to production traffic, you will see your models serve millions of requests per day and watch retailer KPIs climb as a direct result of your work.