Staff Machine Learning Framework/Runtime Software Engineer

In this role you will develop and optimise ML inference runtimes within Arm’s modular codebase, bridging runtime, compiler, driver, and hardware teams. You’ll analyse performance and functional issues across frameworks and hardware abstractions, contributing to end-to-end AI execution from framework models to accelerator-enabled backends. You’ll work across a large codebase to design, test, and integrate solutions that boost efficiency and bring AI capabilities to Arm technology. This is a hands-on role for engineers who love solving complex performance challenges at scale.

• health and wellbeing support
• financial wellbeing
• time away from work
• professional development
• collaboration with global engineering teams

• analyse, investigate, diagnose, and resolve complex functional or performance issues across frameworks, runtimes, compilers, drivers and hardware abstraction layers
• design, develop, integrate, test and optimise solutions within a large, modular codebase
• collaborate across runtime, compiler, driver, and hardware teams to deliver end-to-end AI execution workflows
• support performance optimisation and validation activities across the AI software stack from framework to hardware acceleration
• work on backend integration mechanisms, execution providers, and plugins as part of ML inference workflows

• strong C++ software development experience
• experience with machine learning inference frameworks (LiteRT, TensorFlow Lite, ONNX Runtime) or similar
• understanding of how AI models execute in practice (graph processing, operator execution, memory management, backend integration)
• ability to analyse and resolve complex performance issues across frameworks, runtimes, compilers, drivers, and hardware abstraction layers
• interest in machine learning systems, AI acceleration, and performance optimisation
• collaboration across multiple teams to design and optimise solutions in a large codebase

• strong analytical and problem-solving skills
• interest in machine learning systems and AI acceleration
• cross-functional collaboration
• C++ software development
• machine learning inference frameworks (LiteRT, TensorFlow Lite, ONNX Runtime)
• graph processing and operator execution