CUDA Kernel Developer

JR United Kingdom; Cambridge; Hybrid; GBP 80,000 - 90,000; Full time

Job summary

A multinational semiconductor scale-up in Cambridge is seeking a CUDA Kernel Developer to join their growing team. In this role, you will develop high-performance kernels for ML operators and mentor less experienced engineers. The position offers hybrid working, attractive salary, and potential for career progression in an innovative environment.

Qualifications

• Extensive experience in kernel development for GPUs.
• Experience with OpenCL, CUDA, or similar parallel programming languages.
• Understanding of ML frameworks such as TensorFlow and PyTorch.

Responsibilities

• Develop and optimise high-performance kernels for ML operators on NPU architectures.
• Collaborate with hardware and software teams to integrate kernels into the NPU framework.
• Mentor junior engineers and utilise hardware features of GPUs.

Skills

Kernel development for GPUs

OpenCL

CUDA

TensorFlow

PyTorch

Independent work

Job description

Social network you want to login/join with:

£80,000 - £90,000 + bonus & hybrid working!

I'm currently working with a Cambridge-based, multinational Semiconductor scale-up focused on developing AI accelerators. You will have the opportunity to work in a rapidly changing environment where your ideas will become innovative products, services, and customer experiences. The company is successful and growing, offering career progression and a supportive environment.

They are seeking a CUDA Kernel Developer to develop and optimise high-performance kernels for ML operators on NPU architectures. You will join a talented team of 5 engineers during an exciting growth phase. The role requires experience in mentoring junior engineers and utilising hardware features of GPUs and accelerators tailored for AI applications. Collaboration with hardware and software teams to integrate kernels into the NPU framework is also essential.

Requirements for a successful CUDA Kernel Developer:

• Extensive experience in kernel development for GPUs
• Experience with OpenCL, CUDA, or similar parallel programming languages
• Understanding of ML frameworks such as TensorFlow and PyTorch
• Ability to work independently in a fast-paced environment

If you are a CUDA Kernel Developer seeking a new opportunity within an exciting semiconductor scale-up, please apply to learn more.

For this or other software opportunities across the UK, please contact Jack Bird at IC Resources. [emailprotected]