Job Description:
In the Developer Platforms group at Arm, our mission is to make AI software development on the Arm architecture simple and intuitive. We are growing our team and are looking for a passionate Staff ML Engineer to help us build our expertise in ML tooling and deliver the next generation of developer experiences for Arm.
The ideal candidate will combine strong software engineering skills with deep practical experience of ML model optimisation and deployment. You will help make sophisticated ML optimisation techniques accessible to developers by building tools that analyse, optimize, and prepare models for efficient execution on Arm-based platforms.
You will work across developer tools, ML frameworks and emerging Arm hardware, helping shape both the technical direction and developer experience of our ML tooling.
Responsibilities:
- Join an established engineering team to define, design and deliver production-grade model analysis and optimisation tooling for edge developers, delivered through web services and desktop applications.
- Provide technical leadership in the ML tooling space, identifying gaps in existing technologies and helping shape the architecture and roadmap for our ML developer tooling.
- Work across ML frameworks, runtimes and both existing and future hardware to understand capabilities and constraints and turn these into intuitive experiences for developers who may not be ML experts.
- Build effective relationships with distributed engineering teams, product managers and UX specialists to enable strong collaboration and ensure our tooling meets real developer needs.
- Drive demonstrable engineering quality through maintainable software, automated testing, continuous integration, effective engineering processes and customer feedback.
Required Skills and Experience :
- A strong understanding of neural-network architecture and model execution, including computational graph representation and transformation, operator dispatch, memory planning, backend partitioning and execution across heterogeneous hardware.
- Proven experience in hardware-aware model optimisation techniques such as quantisation, graph optimisation, operator fusionandprecision reduction
- Experience across multiple ML frameworks, model formats and inference runtimes, such as PyTorch, ONNX/ONNX Runtime, ExecuTorch, TensorFlow/LiteRT and OpenVINO.
- Experience analysing, profiling and debugging ML workloads, including model compatibility, performance and the trade-offs between latency, memory, model size and accuracy.
- A passion for building high-quality developer tools and intuitive workflows that make complex ML capabilities intuitive and easy to use by our target developers.
“Nice To Have” Skills and Experience :