Hardware/Software Co-Design Architect

Hardware/Software Co-Design Architect – AI Accelerators Join a VC-backed start-up building the infrastructure that will enable autonomous chip design. Working alongside engineers from NVIDIA, AMD, Intel, Groq and MediaTek, you'll help shape the next generation of AI accelerator architectures in collaboration with some of the world's leading AI labs. About the Role We're looking for a Hardware/Software Co-Design Architect to bridge AI workloads and silicon architecture. You'll analyse frontier inference workloads and translate performance bottlenecks into architectural decisions, influencing everything from memory hierarchy and execution models to hardware/software interfaces and accelerator programmability. What You'll Do Analyse AI inference workloads to identify architectural bottlenecks and optimisation opportunities. Define hardware/software interfaces for next-generation AI accelerators. Influence architectural decisions across memory hierarchy, execution model and data movement. Work closely with compiler, systems software and hardware teams to guide accelerator design from concept through implementation. Help shape the long-term technical direction of a greenfield AI compute platform. What You'll Bring Experience spanning AI systems, hardware/software co-design, accelerator architecture or systems performance. Strong understanding of GPU, NPU or custom accelerator architectures. Experience analysing workload behaviour and translating it into hardware or system architecture decisions. Knowledge of performance trade-offs including latency, bandwidth, utilisation, power and memory movement. Ability to collaborate across architecture, compiler, runtime and hardware engineering teams. Bonus Points Experience influencing ISA, memory hierarchy, execution models or accelerator programming models. Background in compiler or runtime development for custom hardware. Exposure to ML-for-EDA, architectural simulation or AI-assisted chip design. #J-18808-Ljbffr Oho Group