The Computer Vision Engineer at Instrumental will own manufacturing data insight and analysis automation.  He or she will own computer vision, machine learning, and artificial intelligence— collaborating with both hardware and software engineers in the process.

In this role, the Computer Vision Engineer will explore and implement ways to generate insights about our customers’ product and manufacturing processes.  He or she will leverage existing computer vision tools to develop novel image analysis technology that will be run on large (and continuously growing) datasets of images.  To this end, scalability of algorithms, repeatability of measurements, and robustness of data extraction are paramount.  We are looking for someone who can identify powerful applications of computer vision and/or machine learning and implement them in a way the delivers highly accurate and repeatable results for our customers.

What You’ll Do

  • Develop an AI framework to extract useful information from images and other data from assembly lines, which will learn to robustly measure units, detect defects, and more

  • Develop a toolbox of generalized models that can be applied to many different types of units and images without custom tuning, and measure their performance on real data

  • Identify opportunities to use machine learning to improve data quality or the types of analysis we can offer

  • Identify, explore, and implement relevant state-of-the-art computer vision or machine learning methods

  • Quickly prototype and iterate on new algorithmic concepts, and prioritize them based on customer needs

  • Work closely with hardware development engineers to select or improve sensors, and design image capture strategies based on vision algorithm requirements

We’re Looking for Someone Who

  • Loves coding and has a minimum of 2 years of experience in a professional software developer role — we will ask for a code sample of something you’ve previously written in a production language (e.g. Python, Scala, Java, or C/C++)

  • Demonstrated interest and desire in working with OpenCV (beyond Matlab), machine learning, or related fields and libraries - could be at work or with personal projects

  • Has a strong background in computer science with solid fundamentals in algorithms and the math behind them. Computer vision, graphics, machine learning and related fields are a great bonus, but are not strictly necessary.

  • Obsesses over our customers’ needs, specifically accuracy and repeatability

  • Balances attention to detail with the ability to be scrappy and get stuff done

  • Can communicate complex ideas effectively with technical and non-technical people

  • Is good at thinking out loud, difficult to frustrate, and fun to work with

  • Peripheral experiences with mechatronics, metrology, photogrammetry, cameras, machine learning, research in computer vision space, and/or big data is a plus

All candidates must have an unrestricted right to work in the US.

Write algorithms to automatically find defects like the bunched wires highlighted in the unit above (shown next to an ideal unit).

Write algorithms to automatically find defects like the bunched wires highlighted in the unit above (shown next to an ideal unit).