Machine Learning Engineer: Perception
Hybrid (San Francisco, CA)Full-timePosted 3mo ago
About the role
Bedrock is bringing autonomy to the construction industry, deploying production-scale 3D perception systems (fusion of LiDAR and camera data) for heavy equipment with a focus on robustness in real-world, dynamic environments. The role involves designing early fusion architectures, deploying to edge hardware, and collaborating across teams to create state-of-the-art representations for downstream use cases.
Responsibilities
- Design Early Fusion Architectures: Develop and train state-of-the-art models (e.g., BEV-based transformers) that fuse raw Lidar and Camera data to solve for object detection and semantic segmentation.
- Tackle "Messy" Physics: Build perception systems robust enough to handle dynamic occlusion (seeing the robot’s own arm/bucket), particulates (dust, snow, rain), and high-vibration conditions.
- Deploy to the Edge: Optimize models for inference on embedded hardware. Debug system-level issues, such as sensor calibration drift and latency bottlenecks.
- Collaborating with other teams to create state-of-the-art representations for downstream use cases.
Qualifications
- Production ML Experience: 5+ years of experience taking deep learning models from research to real-world production using PyTorch, Tensorflow, or JAX.
- 3D Geometry & Calibration: Deep understanding of SE(3) transformations, homogeneous coordinates, and intrinsic/extrinsic sensor calibration. Ability to project a 3D LiDAR point onto a 2D image pixel.
- Early Fusion Expertise: Practical experience with architectures that fuse modalities at the feature level (e.g., BEVFusion, TransFuser, PointPainting).
- SOTA Object Detection experience with modern transformer-based architectures (DETR, PETR, etc…) including temporal models (PETRv2, StreamPETR).
- Systems Fluency: Proficient in Python; comfortable with C++ or Rust; understanding of memory management and real-time constraints.
- Data Intuition: Understanding that data alignment often beats bigger models; ability to improve data infrastructure for ground-truth quality.
Skills
Required:Production ML3D Geometry & CalibrationEarly Fusion ArchitecturesObject Detection (SOTA)Systems ProgrammingData Quality & Ground Truth
Last updated Sep 10, 2026. Interested? Apply on the company site.
About Bedrock Robotics
Advanced Autonomy for the Built World
Advanced Autonomy for the Built World
AutonomyBedrock Operatorautonomyheavy machineryindustrial automationindustrial roboticsmachine intelligenceon-site autonomy
More open roles at Bedrock Robotics
- Hydraulics System Engineer3d agoHybrid (San Francisco, CA) · Engineering
- Electrical Systems Integration Engineer3d agoHybrid (San Francisco, CA), Hybrid (Lockhart, TX) · Engineering
- Bedrock Training & Certification Lead3d agoHybrid (San Francisco, CA), Hybrid (Lockhart, TX) · Operations
- Cloud - Staff Site Reliability Engineer3d agoHybrid (New York, NY), Hybrid (San Francisco, CA) · Engineering
- Head of Bedrock Operations Center1w agoHybrid (San Francisco, CA), Hybrid (Lockhart, TX) · Operations