Senior GPU Programmer

Department: Technology
Team: Networks
Location: London (Hybrid 2 days a week in the office) Remote can be considered if exceptional circumstances 
Position type: Permanent

What the hiring manager says

You will be working for a world class AI applications and systems team. This is a new team with the opportunity to play a foundational role. We are moving from a successful proof-of-concept to beta-product phase. A key aspect of commercial success relates to the cost and performance characteristics of our systems.The role of AI optimisation engineer is key in ensuring product profitability and competitiveness.  

Dr G.A.McHale, Technical Director, AI & Data Science

About the Team

The team has a strong AI backbone. We are recruiting 2 key members; one will be concentrating on data and infrastructure and the other one will be concentrating on AI algorithm optimisation (this role).

About the Role

The key accountabilities include GPU optimisation of existing algorithms, hardware specific optimisations relating to memory and CPU utilisation. Other work includes hybrid-LLM system optimisation, and possible pre-filtering algorithms for reduced computational loads. 

The success for this role would be significant improvements in the costs of batch processing and inference costs.

About You

• This is a hands-on programming role. Our current phase of development prioritizes on premises computing.

• You must have GPU programming experience, with an interest in broadening your range of expertise to areas associated with AI and natural language processing.

• You will be highly competent in C or C++ and have at least basic competency in Python.

• Suitable candidates from the video Games industry should be able to demonstrable GPU programming experience.
• Candidates are expected to be highly numerate with demonstrable mathematics expertise in at least one domain and an interest in mastering others.
• A background in electronis or hardware is a plus. Any knowledge of Telecommunications, RAN and its supporting hardware would be valuable.
• Experience of or a general knowledge of patents would also be of interest.