Edge AI Perception Engineer for DoD & Autonomy

Decisive Point; Ann Arbor (MI); On-site; USD 140,000 - 230,000; Full time

Job summary

Applied Intuition Defense is seeking a Perception Engineer to advance perception models across maritime, land, and air domains, optimizing for real-time performance on Jetson-like hardware. You will work with transformer-based models, large-scale data, and a team of AI experts to push the boundaries of physical AI used in defense applications.

The role emphasizes hands-on ML, full-stack tooling, and collaboration with cross-functional teams to ensure robust, scalable solutions with DoD testing

Qualifications

• Advanced degree or equivalent practical experience in ML, CS, or engineering.
• Experience training ML models on multi-GPU systems.
• Experience deploying ML models to edge hardware (Jetson/embedded).
• Proficiency in Python and C++.
• Experience with modern AI tooling to accelerate development.

Responsibilities

• Develop perception models with very low error rates on mission-relevant targets.
• Deploy and optimize models on Jetson Orin+ hardware for real-time performance.
• Build large foundation models to support data labeling and labeling tools.
• Create and maintain tools for the perception stack and data generation pipelines.
• Collaborate with hardware, sensor, autonomy and testing teams for DoD deployments.

Skills

Python

C++

Edge deployment

ML model optimization

Education

MS/PhD in CS/CE/Robotics or 5+ yrs exp

Tools

PyTorch

Jetson

GitHub Copilot

Cursor

Job description

Applied Intuition Defense is seeking a Perception Engineer to advance perception models across maritime, land, and air domains, optimizing for real-time performance on Jetson-like hardware. You will work with transformer-based models, large-scale data, and a team of AI experts to push the boundaries of physical AI used in defense applications.

The role emphasizes hands-on ML, full-stack tooling, and collaboration with cross-functional teams to ensure robust, scalable solutions with DoD testing