EngineerJobs.io
← Back to all jobs

Job Description

Build continuously updated 3D online maps for self-driving cars by combining multi-sensor perception with learned PyTorch models and production C++ geometry. This onsite role in Santa Clara, CA supports deployment on NVIDIA DRIVE Orin and Thor, with real-time and safety constraints and close integration across the AV stack.

Salary: USD 184,000 - 356,500 per year. Experience: 5+ years. Education: MS or PhD (or equivalent experience).

What you’ll work on

  • Design, train, and improve PyTorch models that predict online map elements such as lanes, road boundaries, crosswalks, road markings, and topology from camera, LiDAR, and/or radar, using BEV and vectorized-map representations.
  • Tackle difficult scenarios including intersection topology (connectivity between incoming and outgoing lanes, turn paths, merges, and splits) and traffic light understanding (detection, state, and lane association).
  • Implement production C++ for geometry-based post-processing, including curve and surface fitting, association, multi-frame fusion and tracking, topology reasoning, and consistency checks for DNN outputs.
  • Apply multi-view geometry, calibration, coordinate transforms, and sensor fusion to keep 3D maps accurate and stable over time.
  • Optimize and deploy learned models and algorithms alongside geometry on NVIDIA DRIVE Orin and Thor under real-time and safety constraints, then get solutions running in vehicles.
  • Own the data loop: plan datasets, define GT and labeling requirements, partner with labeling and tooling teams to produce high-quality GT, and use data mining plus failure-case analysis to drive improvements.
  • Define critical metrics and build evaluation tooling to quantify quality gains using large-scale real-world data.
  • Coordinate with perception, localization, planning, and infrastructure teams to integrate online mapping into the AV stack across L2++, L3, and L4 programs.
  • Contribute to technical direction and architecture decisions across both learned and geometric components.

Requirements

  • MS or PhD in Computer Science, Electrical Engineering, Robotics, or a related field (or equivalent experience).
  • 5+ years of industry or research experience in autonomous driving, robotics, or computer vision.
  • Strong foundation in 3D geometry: camera models, projective geometry, coordinate transforms, optimization, and curve/surface fitting.
  • Strong C++ skills and experience designing and shipping production-quality, fast, reliable, maintainable algorithm code.
  • Hands-on experience training deep neural networks in PyTorch, including model design, training pipelines, debugging, and performance analysis.
  • Data-driven development experience: building datasets, defining labels/GT, choosing metrics, and improving models by analyzing failures.
  • Excellent problem-solving and communication skills, plus the ability to drive technical direction across teams.

Technologies

PyTorch, C++, BEV, LiDAR, radar, camera, multi-view geometry, CUDA, TensorRT, NVIDIA DRIVE Orin, NVIDIA Thor

Benefits

  • Equity
  • Benefits

Ways to stand out

  • Direct experience with DNN-based online or HD map generation or BEV perception (for example, vectorized map prediction or lane topology learning), or equivalent experience.
  • Experience with intersection understanding, including lane connectivity/topology reasoning in complex junctions, or traffic light detection and lane association.
  • Experience combining learned models with geometric post-processing in a production AV system, plus publications or open-source contributions in 3D vision, mapping, or autonomous driving.
  • Background in sensor fusion, multi-object tracking, or temporal fusion for map elements.
  • Experience deploying DNNs and algorithms on embedded GPUs (CUDA, TensorRT), ideally on DRIVE Orin or Thor, under real-time constraints.

Similar Jobs