ML Engineer (Routing)

Andersen is hiring an ML Engineer (Routing) for a project developing machine learning solutions for real-time routing optimization, ETA prediction, and large-scale geospatial data processing. The customer is a global investment management firm providing tailored investment solutions to institutional and private clients. It combines financial expertise with long-term investment strategies to help clients achieve their objectives while adapting to changing market conditions. The organization focuses on innovation, responsible investing, and operational excellence, continuously enhancing its capabilities to deliver sustainable value and support long-term growth. The project is focused on developing machine learning solutions to improve ETA accuracy and routing quality at a scale. It includes building production-grade ML models, processing large-scale geospatial data, and deploying low-latency inference systems for real-time routing optimization.

Responsibilities: - Designing and building ML models that correct and refine the routing engine's ETA estimates, from gradient-boosted trees through to neural and Transformer-based architectures as data scale grows. - Developing traffic-estimation models that turn large-scale GPS data into road-level speeds and historical-traffic profiles and feed them into the routing engine to produce time-of-day-aware ETAs. - Working on map-matching that snaps noisy GPS data onto the road graph, and the spatial aggregations that make movement and speed data usable for modeling. - Improving ETA calculation, smoothing, and rerouting logic to close the gap between predicted and actual arrival times across changing conditions. - Translating routing and ETA goals into ML objectives with the right proxy metrics and non-functional requirements, including loss formulations where under- and over-prediction carry different costs. - Leading evaluation end-to-end, from offline accuracy and routing-quality metrics to the design of online experiments including shadow tests and interference-aware designs such as switchbacks and prove a change improves accuracy before it ships. - Partnering with backend engineers to take models from prototype to low-latency production serving that meets tight latency and throughput targets. - Partnering with product and operations to turn routing and traffic analysis into concrete features and requirements and help extend ETA capabilities across verticals. - Owning the ML lifecycle in production – serving, monitoring data and concept drift, and building the retraining pipelines that hold quality as traffic patterns, cities, and the map shift.

Requirements: - Experience in Machine Learning for 5+ years, including 3+ years building and deploying deep learning models in production. - Direct experience building regression, forecasting, or other supervised ML systems for real-world prediction problems. - Strong Python skills and hands-on experience with PyTorch, Scikit-learn, Pandas, NumPy, PySpark. - Advanced SQL and distributed data processing experience. - Experience with gradient boosting frameworks (CatBoost, XGBoost, LightGBM). - Ability to design an ML system from scratch in at least one area, including data analysis, processing, and feature engineering through to a model serving in production. - Experience deploying low-latency ML services in production environments. - Experience with MLOps practices, model monitoring, retraining pipelines, and lifecycle management. - Level of English – from Upper-Intermediate and above.

Nice to have: - Subject matter depth in ETA / travel-time prediction, traffic estimation, or routing-engine quality. - Hands-on experience with open-source routing engines (e.g., Valhalla, OSRM, GraphHopper) and concepts such as map-matching, speed profiles, road-graph tiles, and historical traffic. - Experience in mapping, location, or geospatial products. - Experience building for developing markets, where the underlying map and address data is weak. - Experience with cloud data and ML platforms such as BigQuery or Databricks (certifications is a plus), and with distributed deep-learning training.

Compensation: 2800–4550 EUR per month.