Principal Computer Vision Engineer

Catapult Sports•Boston, MA

1d•Hybrid

About The Position

Catapult is a sports technology company that empowers professional teams to make data-driven decisions. We deliver health, performance, video, and AI insights from the locker room to competitive environments, ensuring every decision is an opportunity to gain an advantage, sharpen performance, and build lasting success. We are looking for a Principal Computer Vision & Tracking Engineer to own and drive the computer vision and real-time tracking systems at the heart of our camera-based products. Based in Boston (hybrid), you will lead a small team building the perception stack that turns vision data into athlete performance insights – from neural network inference on edge AI accelerators, to 3D object tracking, to the data science that validates and extends what our system can measure. Today, that means tracking barbell-based weightlifting with sub-rep precision; tomorrow it means general human movement. This is a role for someone who thrives on breadth. In any given week, you might be optimizing a model for edge inference, designing a new movement analysis algorithm, or analyzing data to produce athlete profiles. You'll report to a Senior Director of Engineering and player-coach a small pod of engineers while collaborating closely with our embedded platform and applications teams. If you've built real-time perception systems that had to work reliably in the real world – on real hardware, with real constraints – this is your kind of problem.

Requirements

8+ years of software engineering experience, with significant depth in at least one of: computer vision, real-time systems, ML/AI deployment, or robotics perception.
Strong proficiency in Python and C/C++ – you're comfortable across the full stack from model training to performance-critical native code.
Demonstrated experience shipping ML or CV systems to production, especially on resource-constrained or embedded platforms.
The ability to own ambiguous problems end-to-end: scoping, prototyping, shipping, and measuring.
Comfort with leadership: you've mentored engineers, led technical direction, or driven cross-team initiatives, even if your title didn't say "lead."

Nice To Haves

Deeper knowledge of the mathematics behind tracking systems: 3D geometry, camera models, state estimation, filtering, optimization.
Experience with edge AI accelerators and inference optimization libraries.
Background in pose estimation, human movement analysis, or biomechanics.
Experience with depth cameras (structured light, time-of-flight) and 3D perception.
Familiarity with embedded Linux, Docker on ARM, and systems-level debugging.

Responsibilities

Own the technical direction of our computer vision and tracking domain: the full pipeline from sensor readings through computer vision, motion recognition, and metric calculation.
Lead and mentor a small team (1–2 engineers), setting priorities, reviewing designs, and driving execution while remaining deeply hands-on.
Train, optimize, and deploy neural networks for real-time segmentation and classification on embedded hardware, including model conversion and optimization for edge accelerators.
Extend our tracking and motion recognition pipeline to support new movements and types of analysis.
Own the reliability of the domain: analyze tracking accuracy, build quality metrics, and use data to drive improvements across the pipeline.