About fab2

The fab2 team can build anything. We design all the hardware and software needed to make chips and we make the fabs, tools, and components ourselves.

We're looking for exceptional, hands-on people who can iterate to the limits of what's possible.

fab2 was founded by Sam Zeloof and Jim Keller. Sam is best known for making chips in his garage, and Jim has been a leader in the semiconductor industry for the past 40 years.

About the Team

The tool software group builds Linux control software and microcontroller firmware for fab2’s in-house semiconductor manufacturing equipment. This includes software for sensing, real-time microcontroller firmware, closed loop controls, sequencing, recipe execution, interlocks, operator user interfaces, and diagnostics.

About the Role

You will build computer vision software for lithography and metrology on semiconductor manufacturing equipment. This includes modeling optical systems and developing nanometer-scale alignment (fiducials, calibration methods) through iteration on real hardware. You will also develop vision systems for wafer-handling and process automation equipment. Additionally, you will work on post-lithography metrology, including defect detection and optical proximity correction (OPC) modeling.

You will work closely with full-stack, mechanical, electrical, optical, and process engineers on cross-functional problems.

We are open to a range of experience levels, from early-career to experienced engineers. The most important attribute is a strong mathematical and programming background and the ability to experiment and learn new things.

Responsibilities

Develop classical computer vision algorithms for inspection, measurement/metrology, alignment, positioning, defect detection, and camera calibration.

Build systems that extract reliable measurements and process information from images, including dimensions, feature boundaries, and manufacturing variation.

Develop image registration, camera calibration, and pose estimation capabilities for precision equipment and optical systems.

Apply spatial-domain and frequency-domain image processing techniques to solve scientific and industrial vision problems.

Integrate computer vision systems with cameras, motion systems, robotics, and other electromechanical equipment.

Turn experimental approaches into reliable production software, including testing, validation, performance optimization, and failure handling.

Partner with software, mechanical, electrical, optical, and process engineers to develop new manufacturing and metrology tools.

Required Experience

Bachelor’s or master’s degree in Computer Science, Robotics, Electrical Engineering, Optical Engineering, Applied Physics, or a related field, or equivalent practical experience.

Strong foundation in classical computer vision and image processing.

Experience with techniques such as image registration, feature detection, camera calibration, geometric transformations, pose estimation, or image filtering.

Experience analyzing images in both the spatial and frequency domains.

Understanding of how optics, illumination, sensors, noise, and resolution affect image quality and computer vision performance.

Experience developing computer vision software using Python, C++, OpenCV, or similar tools.

Ability to design experiments, compare technical approaches, and characterize algorithm accuracy and repeatability.

Ability to take computer vision work beyond a prototype and deploy it as maintainable, reliable software.

Nice-to-have

Experience with semiconductor equipment, metrology, manufacturing inspection, microscopy, scientific instrumentation, or industrial automation.

Experience characterizing optical systems using metrics such as resolution, MTF, PSF, etc.

An understanding of optical inaccuracies: chromatic/spherical aberrations, astigmatism, etc.

Experience with precision alignment, motion systems, robotics, or sensor fusion.

Experience working with scientific or industrial cameras.

Research experience, publications, or a strong computer vision project portfolio.