Staff Computer Vision Engineer

Atomic Machines is ushering in a new era in micromanufacturing with its Matter Compiler (MC) technology. The MC enables new classes of micromachines to be designed and built by offering manufacturing processes and a materials library that is inaccessible to semiconductor manufacturing methods. The MC promises to unlock MEMS manufacturing both for the many device classes that never could be made by semiconductor methods but also to open up entirely new classes. Furthermore, the MC is fully digital in the way 3D printing is digital, but where 3D printing produces parts of a single material using a single process, the MC is a multi-process, multi-material technology: bits and raw materials go in and complete, functional micromachines come out.  The Atomic Machines team has also created an exciting first device – one that was only made possible by the existence of the Matter Compiler – that we will be unveiling to the world soon. 
Our offices are in Berkeley and Santa Clara, California.
About This Role:
We are seeking an experienced Staff Engineer to join our growing AI and ModSim team as a Computer Vision (CV) expert. An ideal candidate should have experience developing and deploying CV algorithms for automated manufacturing or robotics systems and the associated quality control. The broad spectrum of CV algorithms goes from object detection to segmentation and classification. You will be required to have an in-depth understanding of CV tools such as OpenCV and deep-learning CV algorithms, and you should feel comfortable driving automated data collection and processing, including curation and labeling. The tools you develop will be a key component of our revolutionary manufacturing platform; as such, you will be required to write robust production level software to be embedded in hardware platforms. You will also have to handle a data-heavy training and testing environment, for which familiarity with ML-Ops tools will be critical. This is an excellent career opportunity for a professional with a proven track record of creating high-quality, robust, and deployable CV algorithms. The ideal candidate is excited to use their skills to help create new classes of devices that can be fabricated and manufactured at scale on our robotics manufacturing platform. They thrive in a highly collaborative workplace, actively engage in the development process, and excel at documenting and presenting their work.

What You'll Do:

• Implement algorithms for automated, task-specific data collection and labeling. 
• Implement algorithms for object detection, segmentation, deblurring, etc., using OpenCV and/or Deep Learning techniques.
• Write robust production code for deploying the above algorithms on hardware platforms.
• Test, benchmark, and optimize algorithms to improve their performance, accuracy, and reliability.
• Handle data-flows for large and noisy datasets for training and testing of deep-learning models.
• Drive the team’s CV efforts in collaboration with platform engineers facilitating the real-time integration of vision in our manufacturing platform.

What You'll Need:

• Strong programming skills in Python, with a strong understanding of and experience in traditional CV algorithms.
• Proficiency in deep learning frameworks (e.g., PyTorch, Tensorflow).
• Experience with creating and using a data collection, annotation, and labeling pipeline.
• Knowledge of ML-Ops fundamentals (release definitions, staging/test environments, cloud deployments).
• Good understanding of the foundations of robotics (affine transformations, hand-eye calibration, etc).
• Strong problem-solving skills, attention to detail, communication, collaboration and presentation skills. 
• Experience working in cross-functional teams. 
• Bachelor’s degree in Computer Science, Software Engineering, Robotics, or a related field, with 7+ years of experience in the industrial application of CV/ML (5+ years for candidates with a Master’s degree or higher).

Bonus Points For:

• Experience with cloud infrastructures (AWS, Azure, etc.) and communication protocols.
• Experience with active learning methods for data collection and/or optimization algorithms for hyperparameter tuning.
• Experience deploying CV algorithms for real-time applications on compute-constrained hardware.