[Please note that this position is open for candidates who are available to start before February 2024.]
Are you excited about being a creator of the next-gen robots? Anyware Robotics is a fast-paced startup backed by world-class VCs. Empowered by the latest progress of AI, we are building the next generation of general-purpose robot which can work flexibly in unstructured environment, helping warehouses and manufacturers to significantly increase efficiency and reduce costs. Our team is expanding rapidly to accelerate development and meet growing customer demands.
Anyware Robotics is seeking highly skilled and motivated Perception Engineers to join our team. As a Perception Engineer, you will be at the forefront of developing cutting-edge machine learning (ML) techniques to solve complex perception problems in mobile manipulators. You will play a key role in enabling our robots to effectively interact with the world around them by improving their scene understanding, localization, object pose estimation, and obstacle detection capabilities.
- Design and implement vision systems for robotics, including calibration, filtering, detection and perception.
- Research and evaluate new ML architectures to solve novel vision problems.
- Train and adapt existing machine learning architectures for visual scene understanding.
- Create real and synthetic training datasets for ML models.
- Guide and participate in ML infrastructure software engineering projects.
- Regularly disseminate research within the team and in wider company meetings.
- Competence in Python programming and popular ML frameworks.
- Experience researching and developing new ML architectures.
- Minimum of 3 years of experience working on ML with robotics applications (academic research or industry).
Additional Desired Experience:
- Experience with large-scale ML training infrastructure.
- Experience training and evaluating ML models for visual scene understanding, such as object detection/instance segmentation, depth estimation/completion, visual odometry/SLAM, and grasp estimation.
- Hands-on experience with robots.
- Competent with C++ programming.
- Practical knowledge of 3D geometry, projective geometry, and camera models.
- Experience with robot simulation software such as Gazebo, MuJoCo, Bullet, Unity, etc.
- Experience or research in visual SLAM.
- Familiarity with typical robotics sensors and calibration procedures, including 3D sensors (ToF, structured light, stereo), RGB cameras, IMUs.
- Experience with real and synthetic dataset capture.
- Experience with computer graphics, such as Blender and procedural generation.
Benefits & Perks
- Competitive compensation & stock
- Comprehensive health insurance
- Flexible PTO
- Regular off-site events