Computer Vision Engineer

Auterion•Manassas, VA

About The Position

Auterion is seeking a Computer Vision Engineer to build and deploy perception systems that operate in real time on embedded computers aboard autonomous aerial platforms. This is a hands-on engineering role focused on real-world deployment rather than offline research, spanning classical computer vision, camera integration and validation, calibration, and optimization for constrained hardware. The engineer will take systems from bench testing to flight, examine captured logs, and improve performance based on field results. The platforms serve commercial and government customers in applications such as inspection, mapping, search and rescue, and security. The role involves implementing real-time perception algorithms for detecting and tracking dynamic objects and estimating their states from a moving platform; integrating and validating camera modules; creating repeatable camera and algorithm evaluation procedures; and improving calibration and sensor synchronization. It also includes profiling and optimizing embedded pipelines to meet latency, throughput, and compute budgets, evaluating or retraining detection models using captured data, and supporting field test campaigns alongside pilots and integrators. The engineer owns deliverables end to end, contributes to camera, lens, and sensor selection, and communicates progress and risks to the team.

Requirements

Degree in Computer Science, Electrical Engineering, Robotics, or a related field, or equivalent practical experience.
Experience building computer vision systems, with exposure to resource-constrained or embedded platforms.
Working knowledge of classical computer vision, including feature detection and matching, optical flow, camera models and calibration, and multiple-view geometry.
Practical experience with cameras as hardware, including sensor and lens behavior, exposure, timing, and camera-module integration issues.
Proficiency in C++ and Python, with practical experience using OpenCV.
Evidence of taking a project from problem definition to a system running on hardware.
Comfort working in the field alongside pilots and using real logs.

Nice To Haves

Camera bring-up below the level of a vendor SDK, including MIPI CSI-2, USB, or GigE, and driver- or V4L2-level work.
Experience with ISP tuning, auto-exposure, or auto-gain pipelines.
Experience evaluating or retraining detection models on real captured data.
Experience with state estimation and sensor fusion, visual inertial odometry, or visual SLAM.
Experience with ROS2, embedded inference runtimes, or aerial platforms such as PX4 or MAVLink-based systems.

Responsibilities

Implement and optimize onboard perception algorithms for real-time detection, temporal association, and state estimation of dynamic objects from a moving platform.
Integrate camera modules onto platforms and validate image quality, timing, latency, and behavior in flight conditions.
Build repeatable test and evaluation procedures to measure camera and algorithm performance and compare results across units, builds, and iterations.
Run and improve camera calibration and sensor synchronization procedures, including multi-camera and multi-sensor setups.
Profile and optimize pipelines for embedded inference against latency, throughput, and compute budgets.
Evaluate or retrain detection models using real captured data.
Support field test campaigns by taking algorithms and cameras from bench to flight, working with pilots and integrators on site, and using captured data to improve systems.
Own deliverables end to end, contribute input to camera, lens, and sensor selection, and keep the team informed about progress and risks.