ML Engineer

Phase 1 (Months 1-3):

Learn our logistics domain and data landscape through pair programming with the tech lead
Build initial data pipelines for multi-source logistics data (AIS vessel tracking, terminal updates, sensor streams)
Prototype exception detection models on real shipment data

Phase 2 (Months 3-9):

Develop unified ML models for delay prediction, discharge timing, and transshipment monitoring
Implement continuous learning pipelines that adapt to evolving shipping patterns
Build explainability layers (SHAP-based) so logistics teams trust the predictions

Phase 3 (Months 9-18):

Scale ML infrastructure for real-time processing (sub-5-minute latency on complex scenarios)
Integrate models into production workflows with A/B testing frameworks
Drive technical decisions on model architecture and data strategy