Robot Perception Engineer - Smart Robotics
TLDR
Develop and optimize visual inspection solutions using computer vision and deep learning, collaborating across teams to deliver high-throughput features for automation.
-
Develop computer vision and deep learning algorithms for visual inspection (defect detection, classification, quality validation) and vision-based navigation (localization, visual servoing, pose estimation)
-
Design data capture strategies, apply augmentation techniques, and train/fine-tune models for inspection and navigation tasks
-
Build and maintain data pipelines and MLOps workflows for training, evaluation, model versioning, and production monitoring
-
Collaborate with Mechanical engineers to design illumination setups and optimize imaging configurations
-
Support model inference optimization for GPU deployment using CUDA, TensorRT, and related frameworks
-
Harden perception solutions for production reliability and work with field teams on deployment and customer rollouts
-
BS or MS in Computer Science, Electrical Engineering, Optics, or a related field with 1–3 years in computer vision/ML
-
Strong Python skills with experience in PyTorch or similar frameworks
-
Familiarity with image acquisition, camera systems, and sensor integration
-
Solid understanding of imaging systems (cameras, sensors, optics, lighting)
-
Familiarity with 3D geometry, pose estimation, and basic electronics for vision systems
-
Experience with GPU inference optimization and industrial camera standards (e.g., GigE Vision, GenICam)
-
Familiarity with camera sensor characteristics (rolling vs global shutter, dynamic range, noise)
-
Exposure to C/C++, MLOps tools, or data annotation workflows
-
Experience with data annotation, labeling workflows, and active learning strategies
-
Familiarity with robotics/vision topics (SLAM, ROS2, sensor fusion) and manufacturing/quality systems
Bright Machines is an innovative manufacturer specializing in AI-powered robotics to streamline the assembly of data center infrastructure. We cater to the needs of hyperscalers and OEMs, leveraging advanced technology to efficiently build vital hardware that fulfills the growing demand for computational power.