JOBSEARCHER

Perception Engineer

Overview: In this role you will advance an AI-forward autonomy stack for contested environments, focusing on vision and multi-sensor perception to enable navigation, targeting, and recognition on Mach products. You will bridge research and real-world deployment, working across deep learning, computer vision, and embedded systems to deliver robust perception solutions. You’ll shape data pipelines, evaluation, and real-time inference on edge hardware. This position offers impact in national security with cutting-edge tech and cross-functional collaboration. Compensation / Benefitsequity offershealth insuranceretirement planspaid time offprofessional developmentcompetitive salary with performance incentives
Responsibilities: Design and refine detection/segmentation/tracking models for EO/IR and multi-spectral imagery; develop large-scale datasets and robust generalizationSet up training/evaluation pipelines and implement continuous regression/testing regimes from field dataOptimize for real-time embedded inference (quantization/pruning, TensorRT/ONNX Runtime) on Jetson-class hardwareFuse vision with radar/LiDAR/RF cues for decision-level fusion and track managementBuild visualization and analysis tools using simulation, HITL, and flight data; plan end-to-end tests with hardware teamsInstrument health metrics, drift detection, and test documentation aligned to performance requirementsConduct simulation-based testing with high-fidelity sensor models and validate with real-world datasets
Key requirements: Production C++ on Linux and Python for ML toolingCUDA backend experience for performance optimization and debuggingProficiency with modern detection/segmentation/tracking models and PyTorch training/fine-tuningExperience building large, diverse datasets, labeling/QA pipelines, augmentation, and reproducible trainingHands-on model compression (INT8/FP16) and runtime optimization for real-time constraintsEO/IR imagery experience with real flight/test data 7+ years of experience with BS/MS/PhD in CS/EE/Robotics or equivalent, with track record shipping ML models to productionCross-functional collaborationStrong problem-solving and attention to detailClear documentation and testing disciplineC++ on LinuxPython for ML toolingCUDA