Our client is a global fabless semiconductor company enabling nearly 2 billion connected devices a year. They are a market leader in developing innovative SoCs for mobile devices, home entertainment, connectivity and IoT products. As a leader in SoC design, the company’s GPU IP team is focused on industry leading, feature rich and PPA competitive GFX IP optimisation, customisation and development, deployed across flagship and mainstream mobile SoCs as well as adjacent markets including laptop, IoT, AI, VR/AR and automotive. The team is growing across Architecture, Driver, Compiler, Performance, Power and Model disciplines at their Cambridge site.
Please note – this role will 100% onsite in the Cambridge area.
What you will be doing
You will bring your GPU driver expertise to a full stack GPU software role, with exposure across the entire stack from industry standard graphics and compute APIs through to kernel mode drivers, GPU firmware and performance optimisation.
Working closely with Architecture, Micro‑architecture, Compiler and Model teams, you will help unlock maximum performance from the silicon and contribute to hardware and software co‑design efforts that ensure seamless integration across every layer.
Your work will span architecting and optimising features for APIs including Vulkan, DirectX, OpenGL ES and OpenCL, working deep in kernel mode drivers and GPU firmware to improve hardware control, stability, scheduling, memory management and execution efficiency, and identifying and eliminating bottlenecks across driver overhead, memory bandwidth, GPU scheduling, compute throughput and power performance tradeoffs.
You will not be restricted to a single layer of the stack; you will lead complex debug efforts, guide design decisions, mentor engineers and help raise the technical bar across the wider GPU driver team, with your software insights directly shaping next generation GPU architecture.
What we’re looking for
- Experience building GPU drivers, graphics systems or low‑level systems software, and be fluent in C/C++ with a comfort working close to hardware.
- At least 7+ years relevant experience.
- Solid expertise in one or more industry standard APIs such as Vulkan, DirectX 12, OpenCL or OpenGL ES, along with a strong understanding of GPU internals including memory hierarchies, command submission, scheduling, synchronisation, pipeline architecture and performance bottlenecks.
- You will enjoy working across boundaries, from API design through to kernel drivers, firmware, compiler interaction and hardware architecture.
Nice to have
A background in GPU performance analysis, firmware, virtualisation, compiler or runtime optimisation for GPUs or accelerators would be an advantage, as would familiarity with ML/AI workloads and DNN operators and their mapping onto GPU or accelerator architectures, and experience collaborating with architects, micro‑architects, compiler and model teams.
#J-18808-Ljbffr…