Sr./Principal Engineer- Autonomous Vehicle Simulation Domain Expert

• Drive technical direction for map-grounded world foundation models, conditioning generative video/world models on map data, drive data, scenario semantics, trajectories, agent behaviours, weather, and lighting.
• Train, fine-tune, and adapt generative models (diffusion, latent video, transformer-based) for driving scenario generation with full ownership of the ML lifecycle: data curation, training, evaluation, and production pipelines.
• Evaluate and extend state-of-the-art models such as NVIDIA Cosmos / Cosmos-Transfer and comparable open-source alternatives for AV training data generation.

Strategic Leadership

• Lead POC initiatives for map-grounded synthetic scenario generation with key technology partners; define measurable success criteria beyond visual realism - focusing on ML training utility, controllability, and sim-to-real transfer.
• Deliver GO / PIVOT / NO-GO recommendations backed by quantitative evidence.

Simulation & Sim-to-Real

• Bridge generative world models with classical simulation stacks (CARLA, NVIDIA Drive Sim, AlpaSim) for physics-grounded scenarios.
• Author and programmatically generate OpenSCENARIO / OpenDRIVE definitions for both classical and generative pipelines.
• Drive sim-to-real strategy: measure domain gap, identify failure modes, and define acceptable thresholds for downstream model training.

Quality Frameworks for Synthetic Training Data

• Define "good enough" synthetic data for AV perception and planning: when photorealism is required, when label consistency suffices, and when controllability matters most.
• Establish validation frameworks combining objective metrics (distribution coverage, label accuracy, FID-style measures, downstream task performance) with expert evaluation protocols.
• Specify sensor fidelity requirements: noise models, lens distortion, and lidar return characteristics.

Technical Collaboration

• Interface with ML research, perception, and planning teams to ensure synthetic data measurably improves real-world model performance.
• Translate business requirements into technical feasibility assessments for product and executive stakeholders.

This role sits at the intersection of deep learning and AV simulation - and we need someone equally grounded in both. You have taken ML models from research into production, navigating real-world constraints, quality thresholds, and safety requirements along the way.

You have experience in:

Deep Learning & Generative Models

• Proven end-to-end model training experience with clear ownership across data, training, evaluation, and iteration.
• Expertise in generative video, world models, or related generative AI - specifically diffusion models, latent video models, and/or transformer-based world models.
• Experience with high-dimensional spatio-temporal data (video, multi-sensor fusion, driving data).
• Strong Python and PyTorch fundamentals; track record of taking ML models from research to production under real-world constraints.

AV Simulation & Scenario Domain

• 5+ years spanning AV simulation, AV perception/ML, or robotics simulation - with meaningful exposure to both simulation platforms and ML model development.
• Hands-on experience with at least one major simulation platform: CARLA, NVIDIA Drive Sim, or equivalent.
• Fluency in OpenDRIVE and OpenSCENARIO: able to author and generate scenario definitions programmatically.
• Understanding of AV testing workflows, ASAM OpenX standards, ISO 34502, and what scenarios stress-test perception and planning systems.

Synthetic Data Quality & Sim-to-Real

• Ability to evaluate synthetic data for distribution diversity, label consistency, edge-case coverage, and downstream task performance.
• Experience with sim-to-real transfer, domain adaptation, or closing the domain gap measurably.
• Clear point of view on trade-offs between photorealism, label accuracy, controllability, and computational efficiency.

This role can be based in Berlin/Frankfurt/Munich/Amsterdam.

HERE is an equal opportunity employer. We evaluate qualified applicants without regard to race, color, age, gender identity, sexual orientation, marital status, parental status, religion, sex, national origin, disability, veteran status, and other legally protected characteristics.

As part of HERE Technologies employment process, candidates will be required to successfully complete a pre-employment screening process. This offer and any related claims are subject to the successful completion of a pre-employment screening. This will involve employment, education, and criminal verification if applicable.

HERE Technologies sits at a unique intersection: we own some of the world's most detailed map and drive data, and we are building generative AI capabilities to turn that spatial intelligence into controllable, high-quality synthetic driving worlds. We are looking for a hybrid profile - someone who combines deep learning expertise in world foundation models, generative video, and transformers with hands-on AV(Autonomous Vehicle) simulation experience. This is not a pure simulation role, nor a pure ML research role. It is the bridge between the two - and that bridge is where HERE's differentiation lives. You understand both how to train and adapt large generative models (think Cosmos, Cosmos-Transfer, diffusion-based video models, latent world models) and how to ground them in map data and scenario semantics so the output is actually useful for training and validating perception and planning stacks.