Staff, Machine Learning Engineer - BEV/Multi-Modal Perception
About this role
The team at Torc Robotics builds software for automated trucks to transform freight movement. This technical leadership position focuses on developing Bird's-Eye View and multi-modal perception models that unify camera, LiDAR, and radar data to provide spatial understanding for autonomous driving.
An experienced machine learning engineer will drive architectural innovation, large-scale model training, and data-driven improvements for perception capabilities. The daily work centers on advancing model robustness, establishing evaluation frameworks, and mentoring engineering colleagues rather than handling downstream feature integration.
What you will do
- Lead the technical roadmap and development for BEV-based perception models spanning detection, segmentation, and scene understanding.
- Design multi-modal architectures that combine camera, LiDAR, radar, and HD map data into unified spatial representations.
- Manage large-scale training workflows encompassing data sampling, augmentation pipelines, distributed training, and hyperparameter tuning.
- Improve model generalization and robustness to handle long-tail conditions like low visibility and rare scene configurations.
- Create evaluation frameworks to measure geometric accuracy, temporal stability, and cross-domain transfer performance.
- Work alongside sensor calibration, mapping, and fusion teams to maintain cohesive perception model interfaces.
- Mentor machine learning engineers by promoting standards in experimentation, code quality, and model validation.
What you need
- Hold 10 or more years of experience working in deep learning for perception, 3D vision, or autonomous systems.
- Possess an M.S. or Ph.D. in Computer Science, Electrical Engineering, Robotics, or a related field, or have equivalent practical experience.
- Demonstrate expertise in BEV modeling, 3D scene understanding, and multi-view fusion.
- Show a strong background in multi-modal sensor fusion, specifically with camera and LiDAR integration.
- Exhibit proficiency in Python alongside deep learning frameworks such as PyTorch or TensorFlow.
- Bring practical experience with large-scale data pipelines, distributed training, and experiment management systems.
- Display leadership capability in driving machine learning model innovation and mentoring technical teams.
Nice to have
- Experience with autonomous driving or robotics perception in production environments
- Experience with MLOps and infrastructure tools like Ray
- Hands-on expertise in BEV-based ML architectures, LiDAR-vision fusion, or spatial-temporal modeling
- Familiarity with 3D labeling, calibration, and sensor simulation pipelines
- Track record of publications or open-source contributions in top-tier venues such as CVPR, ICCV, NeurIPS, ICRA, or CoRL
Benefits
- Competitive compensation package including a bonus component and stock options
- 100% paid medical, dental, and vision premiums for full-time employees
- 401K plan with a 6% employer match
- Flexibility in schedule and generous paid vacation available immediately after the start date
- Company-wide holiday office closures
- AD+D and Life Insurance
Skills and tech stack
Questions candidates ask
What is the remote work policy for this role?
Torc Robotics is open to hiring in Ann Arbor, MI in a hybrid capacity or remote within the United States.
What is the salary range for this position?
The hiring range for this US-based job opening is $215,500 to $258,600 USD. Pay varies based on job-related knowledge, skills, and experience, and the package also includes a corporate bonus and stock option plan.
What kind of technologies will I be working with?
The role involves working with Python, deep learning frameworks like PyTorch or TensorFlow, and infrastructure tools such as Ray. You will build BEV and multi-modal perception models using camera, LiDAR, radar, and HD map data.