We’re looking for a GPU engineer to work closely with our Vision team to help build and maintain the extremely high performance software at the heart of our product, running both on customer, on-premise hardware and in the cloud.
With demonstrable expertise in C++ and experience running software in production, you’ll be primarily implementing and optimising algorithms on the GPU. These will mainly be image processing algorithms, but these days GPUs are everywhere and if we think we can accelerate a task using a GPU, we’ll probably try!
As well as core C++ expertise, experience with CUDA, deep learning, direct-to-GPU networking and I/O, game engine programming and 3D simulation would be helpful - as would a working knowledge of Python. (To be clear, we don’t expect you to tick all those boxes, although if you do, please apply right away!)
Here are some things you might be doing day-to-day:
- Using tools to Identify and profile a bottleneck in the GPU capture pipeline. Based on the results, recommending options for optimisation
- Based on a review of the technical literature, implement a high performance image processing algorithm on the GPU from scratch
- Review existing CUDA research code, and refactor based on current best practices for a production environment
- Recognise an opportunity for GPU parallelism in existing CPU algorithms
- Working with the ML engineers to integrate a deep learning pipeline into the existing CUDA code base
- An inclusive and supportive culture to help you achieve your maximum potential
- Competitive pay and a share option scheme depending on experience
- Flexible working hours and remote first culture (fully supported home working)
- A centrally located office should you wish to use it
- 5% employer match contribution to your pension
- 26 days holiday with buy/sell scheme
- A real commitment to your training & development (including paid courses / subscriptions)