Meet your team
You'll join the ALF (Localization Features) team within TomTom's ADAS & ADS Product Unit, which develops the HD maps and ADAS technology that powers real-time location intelligence for major automotive and tech companies. ALF is responsible for the localization features layer of the HD map and the ML systems that extract them from real-world sensor data at world scale.
You'll work shoulder-to-shoulder with applied scientists and software engineers, partner with adjacent teams, and own the quality trajectory of features that ship to customers under hard product targets.
What you will do:
- Work with a team of medior and senior engineers and applied scientists to develop high-quality algorithms and ML software that powers TomTom's HD maps for ADAS.
- Lead the design, implementation, and integration of algorithms, ML systems, and data pipelines within your area of focus, from problem framing through experimentation to production rollout.
- Drive measurable improvements in output quality (recall, precision, latency, cost) against hard customer-facing targets.
- Lead well-scoped projects and own components within the team's processing pipelines, from upstream input data through algorithmic and ML processing to validated outputs published to downstream consumers.
- Tackle complex technical problems at scale: noisy upstream signals, geospatial geometry, ground truth quality, and large-scale evaluation pipelines.
- Build iteratively using agile methodologies and rigorous experimentation; document outcomes so the team can build on them.
- Mentor junior engineers and interns, provide insightful code reviews, and contribute to hiring as an interviewer.
What you will need:
- 4+ years of professional Applied Science, Machine Learning, algorithm development, or related experience.
- Bachelor's degree (minimum) in Computer Science, Machine Learning, Computer Vision, Geospatial Science, Statistics, or a related quantitative field. Master's or PhD is a plus.
- Solid fundamentals in algorithm design and analysis: data structures, complexity reasoning, and applied algorithms for geospatial and signal-processing problems.
- Solid fundamentals in machine learning: model training and evaluation, statistics, and experimental design.
- Proficiency in Python; experience with at least one ML framework (PyTorch, TensorFlow, or equivalent) and at least one large-scale data processing framework (Spark, Databricks, or equivalent).
- Experience taking algorithms, ML models, or data pipelines into production, not just running experiments offline, and leading well-scoped projects or components to delivery with minimal guidance.
- Experience mentoring junior colleagues and providing insightful code reviews.
- Breadth across applied science: comfortable working across algorithms, modeling, evaluation, ground truth quality, and methodology, with depth in some of them. You should be able to pick up a new sub-domain quickly rather than only operating in a single area of expertise.
- Working familiarity with several of the following (no need to be deep in all): algorithm design for geospatial and geometric problems (polygon geometry, map-matching, spatial indexing), classical ML and clustering on noisy sensor data, computer vision (detection, segmentation), data and ML pipelines at scale (training pipelines, MLOps, dataset generation).
- Proficient in written and verbal communication in English.
- Curiosity and desire to learn, and to expand your skill set across the ML stack.
- Ability to solve complex problems on your own, taking a new perspective on existing solutions and leveraging your experience, peers, and other resources.