Senior Routing Ml Engineer

Bonapolia; Madrid; Hybrid; EUR 70,000 - 95,000; Full time

Benefits offered by this job

Flexible relocation support

Job summary

Bonapolia is seeking an experienced machine learning engineer to design and deploy deep learning models for routing and travel-time prediction. You will build models on large GPS datasets, implement map-matching, and optimize latency for production serving.

You’ll collaborate with backend, product, and operations teams to define features and success metrics. You will own the ML lifecycle, from data analysis to offline/online evaluation and continuous retraining, ensuring drift detection and

Qualifications

• 5+ years of machine learning engineering experience building and deploying deep learning models in production.
• Experience building regression, forecasting, or other supervised machine learning systems for production prediction tasks.
• Expert-level proficiency in Python and its core data science libraries (e.g., PySpark, Pandas, NumPy, Scikit-learn, PyTorch; gradient-boosting libraries such as CatBoost/XGBoost/LightGBM).
• SQL.
• Ability to design an ML system from scratch, including data analysis and processing.
• Experience translating business goals into ML problems with appropriate metrics and non-functional requirements.
• Experience designing and evaluating ML experiments.
• Experience with MLOps tools.
• Experience working with large-scale geospatial and behavioral datasets.
• Experience deploying models to production on ML serving infrastructure and optimizing for latency, and awareness of concept drift and how to detect and manage it.
• Comfort working with large-scale geospatial and behavioral data (e.g., GPS traces, H3 spatial indexing)

Responsibilities

• Design and build machine learning models to improve routing and travel time prediction.
• Develop traffic estimation models using large-scale GPS data.
• Implement map-matching solutions for noisy GPS data.
• Improve travel time calculation, smoothing, and rerouting logic.
• Translate routing objectives into machine learning objectives and evaluation metrics.
• Lead offline and online model evaluation activities.
• Collaborate with backend engineers to deploy low-latency production models.
• Partner with product and operations teams to define new features and requirements.
• Own the production ML lifecycle, including serving, monitoring, drift detection, and retraining pipelines.

Skills

5+ years ML engineering

Regression/Forecasting

Python + core libraries

SQL

ML system design

Translate business goals to ML

ML experiments design

MLOps

Geospatial/behavioral data

GPS + H3 indexing

Education

Academic background in CS/Math

Tools

PySpark

Pandas

NumPy

Scikit-learn

PyTorch

XGBoost

LightGBM

CatBoost

MLOps

Geospatial analytics

Job description

Customer Description:

The customer is a global mobility and urban services platform providing transportation and other on-demand services through a digital marketplace.

Project Description:

The work spans deep learning model development, large-scale geospatial data pipelines, and low-latency production serving. The candidate will measure the impact through offline evaluation metrics and the results of online experiments.

Project Phase:

ongoing

Soft Skills:

• Ability to influence teammates and cross-functional stakeholders effectively.
• Curious and improvement-oriented mindset with a willingness to challenge existing approaches.
• Excellent ability to communicate complex technical findings in a clear and concise manner.

Hard Skills / Must Have:

• 5+ years of machine learning engineering experience building and deploying deep learning models in production.
• Experience building regression, forecasting, or other supervised machine learning systems for production prediction tasks.
• Expert-level proficiency in Python and its core data science libraries (e.g., PySpark, Pandas, NumPy, Scikit-learn, PyTorch; gradient-boosting libraries such as CatBoost/XGBoost/LightGBM).
• SQL.
• Ability to design an ML system from scratch, including data analysis and processing.
• Experience translating business goals into ML problems with appropriate metrics and non-functional requirements.
• Experience designing and evaluating ML experiments.
• Experience with MLOps tools.
• Experience working with large-scale geospatial and behavioral datasets.
• Experience deploying models to production on ML serving infrastructure and optimizing for latency, and awareness of concept drift and how to detect and manage it.
• Comfort working with large-scale geospatial and behavioral data (e.g., GPS traces, H3 spatial indexing)

Hard Skills / Nice to Have (Optional):

• Academic background in Computer Science, Mathematics, or a related discipline.
• Experience with travel time prediction, traffic estimation, or routing quality.
• Experience with open-source routing engines.
• Knowledge of map matching, speed profiles, road graph tiles, and historical traffic.
• Experience with mapping, location, or geospatial products.
• Experience building products for developing markets.
• Experience with cloud data and machine learning platforms.

Responsibilities and Tasks:

• Design and build machine learning models to improve routing and travel time prediction.
• Develop traffic estimation models using large-scale GPS data.
• Implement map-matching solutions for noisy GPS data.
• Improve travel time calculation, smoothing, and rerouting logic.
• Translate routing objectives into machine learning objectives and evaluation metrics.
• Lead offline and online model evaluation activities.
• Collaborate with backend engineers to deploy low-latency production models.
• Partner with product and operations teams to define new features and requirements.
• Own the production ML lifecycle, including serving, monitoring, drift detection, and retraining pipelines.

Technology Stack:

}Python, SQL, PySpark, Pandas, NumPy, Scikit-learn, PyTorch, XGBoost, LightGBM, CatBoost, MLOps, ML Lifecycle Management, Production ML Systems, ML Infrastructure, Geospatial Analytics, GPS Data, H3 Spatial Indexing

Additional notes:

The work will be carried out either on client's equipment or via a virtual workstation. Requires fluent Russian, as part of the team communication is conducted in Russian