Senior Principal Software Architect - AI

At Arm, we are looking for an experienced Senior Principal Software Architect to shape how applications and developer frameworks connect to Arm’s AI acceleration hardware.

Do you enjoy solving complex problems across the boundaries of AI, software and computer architecture? This role offers the opportunity to do exactly that!

Our AI technologies span smartphones, laptops, automotive, IoT, robotics, drones and other edge markets. We work across the complete software and hardware stack. This includes emerging AI models and application use cases, frameworks, runtimes, compilers, compute platforms and accelerators.

The role will help shape the architecture and technical direction of Arm’s AI software portfolio. It will identify future requirements and help developers take full advantage of current and future Arm platforms.

Responsibilities

A key part of the role is technical leadership and collaboration. This includes working with senior engineers and architects to develop, review and approve designs. Constructive challenge, thoughtful feedback and building alignment around important decisions are all essential. The successful candidate will act as a trusted technical authority, helping guide decisions across teams and organisational boundaries.

The role also works across hardware and software teams. We aim to achieve effective hardware/software co-design and appropriate reuse across different stacks and markets.

The role will also involve working with senior technologists and technical experts at Arm’s customers and ecosystem partners. These relationships help us understand emerging requirements, exchange technical ideas and build alignment on future technologies.

Other responsibilities include:

• Understanding emerging AI models, workloads and application use cases, and translating these into future software and hardware requirements.
• Working across ecosystems including PyTorch, ExecuTorch, LiteRT and ONNX, plus associated compilers, runtimes and kernel libraries.
• Evaluating emerging technologies and industry trends to understand their implications for Arm., Arm’s approach to hybrid working is designed to create a working environment that supports both high performance and personal wellbeing. We believe in bringing people together face to face to enable us to work at pace, whilst recognizing the value of flexibility. Within that framework, we empower groups/teams to determine their own hybrid working patterns, depending on the work and the team’s needs. Details of what this means for each role will be shared upon application. In some cases, the flexibility we can offer is limited by local legal, regulatory, tax, or other considerations, and where this is the case, we will collaborate with you to find the best solution. Please talk to us to find out more about what this could look like for you.

We are looking for extensive experience in software architecture, systems software or hardware/software co-design.

Deep knowledge of computer and system architecture is essential. This should include processors, memory systems, heterogeneous computing and hardware acceleration.

Strong knowledge of modern AI models and workloads is also required. Candidates should understand the computational characteristics that determine performance on hardware. Familiarity with PyTorch, ExecuTorch, LiteRT, ONNX or ONNX Runtime is important., * Understanding of how AI workloads move through graphs, compilers, runtimes and kernels to hardware.

• Strong C/C++ skills and the ability to engage directly with software implementations.
• Excellent technical judgement when navigating complex and ambiguous problems. This means identifying what matters and bringing structure to uncertainty.
• A strong professional network across the AI, semiconductor or software ecosystem. Experience building trusted relationships with customers, partners and industry peers is important.
• Excellent communication and interpersonal skills. Listening, constructive challenge, consensus building and influencing across organisational boundaries are central to this role.

“Nice To Have” Skills and Experience

We would particularly value experience with NPUs, GPUs, DSPs or other hardware accelerators. Knowledge of ML compiler technologies such as MLIR and LLVM would also be valuable.

Other useful experience includes low-precision inference, LLMs and multimodal models, Android or Linux AI ecosystems, and embedded or edge computing. Experience working with open-source communities is also welcome.