Bedrock Robotics
Machine Learning Engineer: Perception
San Francisco, CA · Hybrid · Full-time · Posted May 4, 2026
- Experience
- 5+ years; no alternatives stated.
About the job
Build and deploy production 3D perception systems that fuse lidar and camera data for autonomous construction equipment.
What you'll do
- Design and train early-fusion models for object detection and semantic segmentation.
- Develop robust perception systems for challenging job-site conditions and optimize them for embedded hardware.
- Debug calibration and latency issues and collaborate on representations for downstream applications.
What they ask for
- 5+ years of taking deep learning models from research to real-world production using PyTorch, Tensorflow, or JAX.
- Deep understanding of 3D geometry, SE(3) transformations, homogeneous coordinates, and sensor calibration.
- Practical experience with feature-level multimodal fusion and modern transformer-based object detection, including temporal models.
- Expert Python skills and ability to read and write C++ or Rust, with understanding of memory management and real-time constraints.
Nice to have
- Experience with occupancy grids, NeRFs, or voxel-based terrain representations.
- Research published at ICRA, IROS, CVPR, ECCV, ICCV, CoRL, or RSS.
Tools
PyTorch, Tensorflow, JAX, Python, C++, Rust, BEVFusion, TransFuser
AI focus
Develop and deploy deep learning perception models that fuse lidar and camera data for object detection and semantic segmentation on autonomous construction equipment.
Model-written summary · the original posting is authoritative