Defense Perception Engineer: Edge ML & Real-Time Vision

Speedrun Talent Network; Sunnyvale (CA); On-site; USD 180,000 - 260,000; Full time

Job summary

Applied Intuition Defense is seeking a Perception Engineer to rapidly advance perception model performance across maritime, land, air, and surface domains using EO/IR data. You will build detector/classifier architectures, develop MLOps tooling, and manage labeling on a large scale with frontier AI.

Expected to collaborate with hardware, sensor, autonomy, and testing teams, deploy to edge devices, and contribute to on-site DoD demonstrations.

Qualifications

• MS or PhD in Computer Engineering, Robotic Engineering, CS or equivalent, or 5+ years of relevant experience.
• Proficiency in training ML models in PyTorch on multi-machine, multi-GPU systems.
• Experience deploying ML models to edge devices and real hardware.
• Strong Python and C++ coding skills for production systems.
• Experience with modern AI development tools to accelerate robust systems.
• Ability to learn new software with minimal supervision

Responsibilities

• Develop perception models with very low error rates for targets.
• Deploy models to Jetson Orin+ boards and optimize real-time performance.
• Build large foundation models for data labeling workflows.
• Create tooling across the perception stack, including labeling and simulation.
• Collaborate with defense, autonomy, and testing teams for on-site DoD demos.
• Scale datasets with simulation, diffusion, and AI data generation

Skills

Python

C++

PyTorch

Multi-machine training

Transformer models

MLOps tooling

Edge deployment

GitHub Copilot

Cursor

Education

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

Tools

GitHub Copilot

Cursor

Unreal Engine-like simulation tools

Job description

Applied Intuition Defense is seeking a Perception Engineer to rapidly advance perception model performance across maritime, land, air, and surface domains using EO/IR data. You will build detector/classifier architectures, develop MLOps tooling, and manage labeling on a large scale with frontier AI.

Expected to collaborate with hardware, sensor, autonomy, and testing teams, deploy to edge devices, and contribute to on-site DoD demonstrations.