Perception Engineer, Machine Learning

Anduril Industries is a defense technology company with a mission to transform U.S. and allied military capabilities with advanced technology. By bringing the expertise, technology, and business model of the 21st century's most innovative companies to the defense industry, Anduril is changing how military systems are designed, built and sold. Anduril's family of systems is powered by Lattice OS, an AI-powered operating system that turns thousands of data streams into a realtime, 3D command and control center. As the world enters an era of strategic competition, Anduril is committed to bringing cutting-edge autonomy, AI, computer vision, sensor fusion, and networking technology to the military in months, not years.

About the Role

We are seeking a Machine Learning Perception Engineer to join our Perception team in Seattle, where we develop production perception systems for autonomous Unmanned Aerial Vehicles (UAVs), Unmanned Surface Vessels (USVs), and Unmanned Ground Vehicles (UGVs) operating across land, sea, and air. Our team is small, agile, and equal parts research and deployment, we push state-of-the-art perception models into the field across multiple business lines and demanding mission environments .

This is a hands-on role for an individual who is fluent in current research trends and equally focused on what it takes to make models maintainable, observable, and reliable once they're deployed on hardware. You will own perception capabilities end-to-end: from data preparation and model design through deployment, monitoring for performance drift in the field, and continuous improvement. Roughly half your time will be spent on new model development and half on hardening and improving existing pipelines.

If you are passionate about tackling complex technical challenges in computer vision and perception and want to see your work flying, sailing, and driving in the real world, we encourage you to apply.

What You'll Do

• Own perception capabilities end-to-end including detection, classification, segmentation, depth estimation, scene description, tracking, sensor fusion, and calibration across our UAV, USV, and UGV platforms .
• Bring state-of-the-art models into production, survey, prototype, and integrate the latest research while ensuring deployments are maintainable and observable in the field .
• Optimize and deploy models to embedded hardware such as NVIDIA Jetson/Orin-class platforms, meeting real-time latency budgets on the order of 30 ms per frame .
• Architect and improve our CI/CD pipelines. We use systems to continuously ingest data, label, and train new models. You would have a direct role in shaping these systems.
• Onboard new sensor modalities and features end-to-end, from cradle to grave, whether that's adding OCR metadata extraction to our pipeline or integrating a new LiDAR point cloud source. Example sensors includes EO, IR, stereo, radar, and LiDAR .
• Train models in the cloud and on-prem, deploy to edge hardware, and maintain a roughly 50/50 split between edge and cloud inference workloads .
• Monitor deployed models for performance divergence and build the tooling needed to detect and respond to drift in the field .
• Collaborate across teams with hardware, systems, and program engineers to ship perception capabilities into production systems.

Required Qualifications

• Education: Bachelor's degree in Computer Science, Electrical Engineering, Robotics, Machine Learning, or a related field. Master's degree or higher preferred.
• Experience: 5+ years of industry experience building and deploying perception or computer vision systems in production.
• Robotics experience, you have shipped perception software onto real robotic platforms (drones, ground vehicles, marine vehicles, manipulators, or similar) .
• Strong proficiency in Python for model development and training, plus production experience in C++ and/or Rust for real-time, on-hardware deployment .
• Hands-on experience with PyTorch and OpenCV. TensorFlow experience is also welcome.
• Demonstrated track record of designing, refining, and deploying ML models into production, including ownership of maintainability and field monitoring.
• Model optimization for embedded inference, quantization, pruning, TensorRT, and ONNX .
• Experience with classical computer vision including camera calibration, homography, multi-view geometry, and feature-based methods.
• Experience with structured and unstructured data and integrating new sensor modalities.
• Familiarity with CI/CD practices and writing and maintaining automated tests around ML pipelines.
• End-to-end ownership mindset, you maintain what you build and design it to be maintainable .
• Cross-team collaboration skills
• U.S. Citizen and eligible to obtain and maintain a U.S. security clearance .

Preferred Qualifications

• Master's or PhD in Computer Vision, Machine Learning, Robotics, or a related field.
• Experience with SLAM
• Experience with AWS training infrastructure, on-prem GPU clusters, and tools such as Weights & Biases, Flyte, and RunAI .
• Experience with synthetic data generation and sim-to-real workflows .
• CUDA / GPU programming experience .
• Background in defense, aerospace, or autonomous systems .
• Background in robotics perception for drones or marine vehicles .
• Experience with real-time embedded vision systems .
• Leadership instincts, comfortable mentoring teammates or contributing to architecture and roadmap conversations.
• Publications or contributions to computer vision research or open-source projects .

Logistics

• Location & Work Model: On-site in Seattle, WA. Occasional travel to events and to our Costa Mesa, CA headquarters several times per year.
• Employment Type: Full-time.
• Clearance: U.S. Citizenship required; must be eligible to obtain and maintain a U.S. security clearance .