5+ years of experience in building and deploying perception systems on physical hardware.
Proficient in Python and C++ for production ML engineering, including model development and deployment.
Hands-on experience with real-time computer vision for detection and tracking from live sensor data.
Experience deploying models on-device using TensorRT, ONNX, or similar tools, with a focus on embedded systems.
Recent background in computer vision, perception, or robotics software related to machine learning for physical systems.
Familiarity with multi-sensor fusion techniques and EO/IR or thermal imagery is a plus.
Experience in counter-UAS, aerial tracking, or working in adversarial environments is advantageous.
Responsibilities
Design and manage the perception architecture for detection, tracking, classification, and sensor fusion.
Develop machine learning models and training pipelines for deployment on embedded devices.
Fuse electro-optical/infrared (EO/IR) and other sensor data within the targeting system.
Optimize perception software for real-time performance under strict latency requirements.
Collaborate with robotics, controls, and hardware teams to ensure perception aligns with system constraints.
Mentor and guide perception engineers as the team expands.
Benefits
On-site work environment in San Francisco, California, five days a week.
Opportunities for professional growth and mentorship within a specialized team.
Engagement in cutting-edge defense technology projects.
Collaboration with multidisciplinary teams in robotics and hardware engineering.
Full Job Description
About the Role
Lead the perception stack for an autonomous defense technology system, owning detection, tracking, classification, and sensor fusion. You will work closely with robotics software, controls, and hardware engineers to deliver perception that performs reliably in real-world conditions. What You'll Do
Design and own perception architecture across detection, tracking, classification, and sensor fusion.
Develop ML models and training pipelines, then deploy them on embedded devices.
Fuse EO/IR and other sensor data across the targeting system.
Optimize perception software for real-time performance on embedded hardware and within tight latency budgets.
Partner with robotics, controls, and hardware teams to align perception capabilities with system constraints.
Mentor perception engineers as the team grows.
What We're Looking For
At least 5 years building and shipping perception systems on physical hardware.
Production ML engineering experience in Python and C++, including model development, training pipelines, and deployment.
Hands-on experience building real-time computer vision detection and tracking from live sensor data, and shipping systems for real-world targets.
Experience deploying models on-device with TensorRT, ONNX, or equivalent, and optimizing for embedded compute and latency constraints.
A recent background in computer vision, perception, robotics software, or related machine learning for physical systems.
Experience with multi-sensor fusion, EO/IR or thermal imagery, moving-camera compensation, or small and long-range targets is valuable.
Counter-UAS, aerial tracking, radar or RF perception, classified programs, and work in adversarial or denied environments are also relevant.
Compensation & Benefits
Base salary: $190,000 to $222,000 annually. Visa sponsorship is not available. Location
On-site in San Francisco, California, United States, five days per week, with range days.
