Senior Applied Data Scientist

1w1 week ago

HERE Technologies

US · Full-time · $140,000 – $160,000

About this role

We are looking for a Senior Applied Data Scientist to drive the development of large-scale computer vision and machine learning solutions. These solutions automatically extract high-fidelity map features from aerial and satellite imagery. Leverage vision foundation models, deep learning, and geospatial analytics to power HERE’s global mapmaking products.

Work on cutting-edge problems like automated extraction of road and lane boundaries and buildings at planetary scale. Handle diverse imagery sources and conditions with production systems. This role sits at the intersection of AI research and production, impacting next-generation maps used by millions worldwide.

Build scalable ML pipelines processing large volumes of satellite imagery with high accuracy and efficiency. Integrate outputs into HERE’s mapmaking systems, aligning with cartographic standards. Address geospatial challenges including variable resolution, diverse sensors, and regional differences.

Define evaluation metrics, conduct experiments, and drive continuous model improvement. Collaborate cross-functionally with data engineering, mapping, and product teams. Translate business needs into ML solutions for real-world geospatial impact.

Requirements

Master’s or PhD in Computer Science, Machine Learning, Data Science, Geomatics, Remote Sensing, or related field (or equivalent practical experience)
Hands-on experience applying machine learning or computer vision to real-world production problems
Strong expertise in deep learning for images, including CNNs, transformers, and large vision models
Practical experience working with satellite or aerial imagery, geospatial data, or remote sensing datasets
Proficiency in Python and major ML frameworks (e.g., PyTorch, TensorFlow)
Experience designing experiments, evaluating model quality, and driving continuous improvement in deployed systems
Experience with vision foundation models or large-scale pre-trained models for imagery (preferred)
Knowledge of geospatial data formats, coordinate systems, and spatial analysis (preferred)

Responsibilities

Design, develop, and deploy computer vision and machine learning models to automatically extract map-observable features from satellite imagery
Leverage and adapt vision foundation models and modern deep learning techniques for geospatial feature extraction at global scale
Build scalable, production-ready ML pipelines that process large volumes of satellite imagery with high accuracy and efficiency
Integrate ML outputs into HERE’s mapmaking systems, ensuring alignment with cartographic quality, consistency, and accuracy standards
Address real-world geospatial challenges such as variable imagery resolution, diverse sensors, regional differences, and environmental conditions
Define evaluation metrics, conduct experiments, and drive continuous improvement of model performance and map data quality
Collaborate cross-functionally with data engineering, mapping, and product teams to translate business needs into ML solutions