Lead Data Scientist

HERE Technologies•United States Home Office,

4d•$160,000 - $170,000•Remote

About The Position

We are hiring a Lead Data Scientist to own applied AI quality, data strategy, and downstream usefulness across advanced AI, computer vision, and perception systems. This person will define how we determine whether a system is working, where it is failing, what quality bar is required for release, and how data, evaluation, and applied modeling should evolve to improve the product. This is a senior role for someone who can bring rigor to ambiguous technical programs, establish evaluation systems, and translate model behavior into product decisions, data strategy, and concrete improvement loops.

Requirements

Strong background in applied machine learning, computer vision, synthetic-data evaluation, or perception-system validation
Experience designing metrics and evaluation frameworks for generative, simulation, or perception systems
Experience connecting model behavior, data quality, and product outcomes in ambiguous AI systems
Ability to translate research-quality experiments into practical engineering and release decisions
Strong analytical judgment and clear written communication
Comfort owning both strategy and execution in a small team
Master’s or PhD in Computer Science, AI, Machine Learning, or related field.
5-8 years of experience in deep learning, computer vision, or multimodal AI.

Nice To Haves

Experience with simulation, autonomous systems, geospatial AI, or map-grounded perception tasks
Familiarity with video quality metrics, structural similarity measures, temporal consistency checks, segmentation and detection evaluation, or label-quality assessment
Experience assessing synthetic-to-real transfer, dataset usefulness for downstream models, data curation strategy, or production quality governance

Responsibilities

Own the evaluation framework and quality strategy for advanced AI and vision systems
Define pass/fail metrics for output quality, structural fidelity, temporal consistency, label quality, robustness, and operational repeatability
Own data and validation strategy for improving model quality and downstream usefulness
Lead artifact auditing, failure taxonomy development, release-quality reporting, and evidence-based prioritization
Measure whether outputs are suitable for perception, mapping, generative AI, and customer-facing use cases
Partner with model, simulation, and platform owners to drive quality improvements and production-readiness decisions
Build and evolve metric suites for output quality, fidelity, repeatability, and downstream usefulness
Define human-review protocols and product acceptance thresholds for complex AI systems
Evaluate whether outputs preserve the structure, semantics, and consistency expected by downstream applications
Translate evaluation findings into data strategy, experiment priorities, and applied modeling opportunities
Help define dataset design, validation slices, and quality-improvement loops across the product
Create experiment and release reports that turn technical output into clear product decisions
Help prioritize what the team should fix next based on evidence rather than intuition
Establish evaluation foundations that remain useful across future AI, perception, and mapping capabilities