About this role
Join a team building portable AI inference infrastructure to enable reliable customer workloads across K3 CPU/NPU environments, with an emphasis on equivalence and safe handling of unsupported models and operators.
Role details:
- Fixed-term Contractor from 15 Oct – 14 Dec 2026.
- Remote / overseas applicants are welcome (global candidates).
- Minimum experience required: 5+ years in ML systems / inference engineering.
What you'll work on:
- AI model intake, conversion and deployment
- K3 NPU compiler/runtime and supported operators
- ONNX, PyTorch export & model runtimes
- Quantisation, calibration & numerical validation
- Edge/accelerator NPU and heterogeneous inference
- Python/C++ and Linux deployment pipelines
- Model equivalence testing and production acceptance
Ideal candidate:
- Strong hands-on experience with ML inference systems
- Experience with model compilers and runtimes
- Familiarity with ONNX and PyTorch workflows
- Practical experience in quantisation and production deployment