Computer Vision Software Engineer, Software Defined Weapon Optic

Anduril • Costa Mesa, CA

4d • $132,000 - $198,000

About The Position

The Computer Vision Software Engineer will develop and optimize computer vision algorithms for a software-defined weapon optic that provides precision aiming and targeting, including counter-UAS applications. The role focuses on solving real-time vision problems such as target detection, tracking, camera and sensor calibration, and geometric pose estimation on resource-constrained embedded platforms.

The engineer will work on a high-performing team within Anduril Warfighter Systems that is building next-generation precision engagement systems. The team combines computer vision, sensor data, and onboard processing to improve first-round accuracy, with algorithms designed to operate within tight compute and power budgets. The position involves collaboration with firmware, mechanical engineering, and manufacturing teams to integrate vision software into production weapon-sight hardware.

Requirements

• Bachelor of Science in Robotics, Computer Science, Electrical Engineering, Mechanical Engineering, Mechatronics, or a related field.
• Proficiency in C++ and Python.
• Strong understanding of fundamental computer vision concepts, including camera projection models, image distortion, rigid transformations, and multi-view geometry.
• Prior academic or professional experience with geometric computer vision, such as SLAM, camera calibration, stereo vision, or photogrammetry.
• Experience developing or deploying algorithms in resource-constrained or embedded environments.
• Good communication skills for collaboration across hardware, software, firmware, and manufacturing teams.
• Eligible to obtain a security clearance.

Nice To Haves

• Master's or PhD in Robotics, Computer Science, Electrical Engineering, Mechanical Engineering, Mechatronics, or a related field.
• Experience with industrial computer vision systems, counter-UAS applications, or precision targeting systems.
• Hands-on experience with Qualcomm processors or platforms, such as Snapdragon, or similar embedded vision SoCs.
• Experience with multi-camera and multi-sensor EO/IR systems and associated calibration techniques.
• Proficiency with mathematical or optimization packages in C++ and Python, such as Ceres, Eigen, OpenCV, NumPy, or SciPy.
• Experience training, optimizing, and deploying detection or tracking models with TensorRT, SNPE, ONNX Runtime, or similar inference frameworks for edge deployment.
• Familiarity with real-time operating systems or embedded Linux environments.
• Familiarity with numeric computing platforms such as MATLAB, Octave, or Jupyter.

Responsibilities

• Design and develop real-time target detection, classification, and tracking models and algorithms optimized for low-compute embedded environments.
• Build and refine offline and online camera and sensor calibration pipelines for multi-sensor weapon sight assemblies, including thermal and electro-optical sensors.
• Develop pose-estimation and localization algorithms using techniques such as bundle adjustment, feature detection and matching, homography fitting, and multi-view geometry.
• Profile, optimize, and deploy vision models and algorithms on constrained hardware, balancing accuracy, latency, and power consumption.
• Create testing and evaluation frameworks, design experiments, analyze field and production data, and establish metrics for detection, tracking, and overall algorithm performance under operational conditions.
• Collaborate with firmware, mechanical engineering, and manufacturing teams to integrate vision software into production weapon-sight hardware.

Benefits

• Highly competitive equity grants are included in the majority of full-time offers and are part of Anduril's total compensation package.
• Top-tier benefits for full-time employees, described as comprehensive and competitive and available at little to no cost to employees.

What This Job Offers

• Job Type: Full-time
• Career Level: Mid Level