Data Scientist

📋 Description Own feature engineering roadmap for ETA & Destination Forecast across 4 commodities. Propose and implement features as dbt models using Airflow for pipelines. Design and run experiments with kpler-ml; log with MLflow; compare to prod baseline. Collaborate with market analysts and product stakeholders to prioritise experiments. Contribute to drift monitoring; validate PSI/KS thresholds using MLflow. Document experiment decisions in MLflow and Confluence. 🎯 Requirements 2+ years ML in production environments, not just research. Experience with geospatial or sequential data (vessel trajectories, H3/S2). Python proficiency to implement features, write dbt models, and script experiments. Familiarity with MLflow or alternative trackers (Weights & Biases, Neptune). Nice to have: maritime domain knowledge; Redshift or columnar warehouses with dbt. 🎁 Benefits Hybrid work arrangement with flexibility. Global team of 850+ experts from 69 countries. Work on production ML for cargo intelligence used by analysts and traders. Inclusive, collaborative culture with strong support for growth.