Representative role.This is a composite of searches we run regularly at this level and stack. If your background matches, apply anyway — we’ll match you to similar live briefs.
An established Adelaide defence prime is hiring mid-to-senior computer vision engineers to work on detection, tracking, and classification from EO/IR imagery. Cleared environment, long-running programmes, real hardware in real conditions.
You won't be chasing SOTA on public benchmarks. You will be closing the gap between model performance and mission performance — out-of-distribution robustness, adversarial conditions, edge deployment.
What you’ll do
- Design, train, and evaluate CV models (detection, segmentation, tracking) on EO/IR and SAR imagery
- Handle real-world data pipeline challenges: labelling, curation, domain adaptation, synthetic augmentation
- Deploy models to edge hardware in cleared environments; work with systems engineers on integration
- Contribute to the ML methodology the team uses — experiment tracking, reproducibility, eval harnesses
What you bring
- 4+ years in computer vision with production deployment experience
- Strong PyTorch or TensorFlow; ONNX export workflow familiar
- Understanding of common failure modes in vision systems (distribution shift, adversarial, occlusion)
- NV1 clearance or demonstrable pathway (Australian citizen, clean background)
Nice-to-haves
- Prior defence or aerospace experience
- SAR / radar imagery background
- Experience with Jetson, Hailo, or similar edge-AI hardware