Perception / Computer Vision Software Engineer

Figure, an AI Robotics company based in Sunnyvale, CA, is seeking Perception/Computer Vision Software Engineers to develop machine learning algorithms and deep neural network architectures for object identification, object pose estimation, scene reconstruction, and geometric shape completion. The ideal candidate should have experience implementing, testing, and deploying perception solutions in C++ and/or Python on real robots, extensive experience with deep learning frameworks and tools, and possess both a theoretical understanding and practical experience with robot perception and computer vision algorithms. The role requires collaboration with other Figure team members to select sensors and determine their placement and to develop and implement a full autonomy stack. Bonus qualifications include experience with deploying real-world perception solutions for commercial applications, experience with GPU programming, and a publication record of developing and deploying perception solutions.

• Design, implement, test, and deploy machine learning algorithms and deep neural network architectures for object identification, object pose estimation, scene reconstruction and/or geometric shape completion
• Train deep learning models using real and digital twin data to achieve robust perception in challenging conditions, such as poor lighting and cluttered scenes
• Evaluate potential perception solutions and make design trade offs and decisions based on robot requirements
• Develop and use modern software engineering techniques to implement high quality, well-tested software
• Engineer and ship high quality, reliable software that will be deployed to the real world
• Collaborate with other Figure team members to select sensors and determine their placement and to develop and implement a full autonomy stack
• Work with the Figure Integration and Test team to evaluate and continuously improve your perception software

• Experience implementing, testing, and deploying perception solutions in C++ and/or Python on real robots
• Extensive experience with deep learning frameworks and tools, such as TensorFLow, PyTorch, TensorRt, and/or ONNX
• Experience training and deploying neural networks on real hardware
• Capable of quickly writing massive amounts of high quality, well-tested, perception software
• Possess both a theoretical understanding and have practical experience with robot perception and computer vision algorithms
• Have a deep knowledge of state of the art techniques, data structures, and software tools
• Thrive in a high pace environment, where solutions are often unclear and require exploration
• Experience in robot perception