Defense Perception Engineer — Edge ML & Real-Time Vision

Applied Intuition; Sunnyvale (CA); On-site; USD 160,000 - 220,000; Full time

Job summary

Applied Intuition Defense seeks a Perception Engineer to advance end-to-end perception systems across maritime, land, air-to-surface and more, deploying models to edge devices and integrating with DoD testing. The role emphasizes hands-on development, large-scale data handling, and collaboration with elite AI talent from Applied Intuition’s self-driving division.

The candidate will work with transformer-based models, vision-language and diffusion techniques, building tools and pipelines for

Qualifications

• Proficiency in training ML models in PyTorch on multi-machine, multi-GPU systems.
• Experience deploying ML models to edge devices.
• Strong coding in Python and/or C++.

Responsibilities

• Develop perception models that find and classify targets with high accuracy.
• Deploy perception models to Jetson Orin+ boards and optimize real-time performance.
• Build tooling across the perception stack: labeling tools, simulation.
• Collaborate with AI talent across Applied Intuition Defense and self-driving teams.
• Scale datasets using simulation, diffusion, and generative AI techniques.
• Create optimized inference software for high FPS on Jetson-like hardware.
• Coordinate with hardware, sensor, and autonomy teams for DoD testing deployments.

Skills

PyTorch

Python

C++

Edge deployment

Multi-machine multi-GPU

Transformer models

Education

MS or PhD in Computer Engineering, Robotic Engineering, Computer Science

Tools

GitHub Copilot

Cursor

Job description

Applied Intuition Defense seeks a Perception Engineer to advance end-to-end perception systems across maritime, land, air-to-surface and more, deploying models to edge devices and integrating with DoD testing. The role emphasizes hands-on development, large-scale data handling, and collaboration with elite AI talent from Applied Intuition’s self-driving division.

The candidate will work with transformer-based models, vision-language and diffusion techniques, building tools and pipelines for