Senior AI Infrastructure Performance Engineer

Pure Resourcing Solutions; Cambridge; Hybrid; GBP 90,000 - 120,000; Full time

Job summary

Pure Resourcing Solutions is seeking a Senior Performance Engineer to join a Cambridge-based AI infrastructure initiative. The role is hybrid, positioned between research and engineering to convert live training metrics into practical optimization insights.

You will create models and calculators forecasting performance, assess accelerator value, and guide architecture decisions that influence procurement and deployment across the organisation and its members.

Qualifications

• A degree in computer science, mathematics, or something adjacent.
• A track record of building performance models or calculators (Python or spreadsheet-based) that forecast system behavior.
• Hands-on GPU/accelerator code optimisation, CUDA or similar.
• Genuine understanding of how LLMs and deep learning models run on real hardware, training vs inference, matrix multiplication, KV-caching.
• Comfortable in profiling tools like Nsight or PyTorch Profiler, and monitoring stacks like Prometheus and Grafana.
• Python for data work, Pandas and NumPy, plus general scripting.

Responsibilities

• Sit between research and engineering teams, pulling real numbers from live training and inference runs.
• Build models and calculators that turn numbers into actionable answers about optimisation and cost.
• Influence purchasing decisions and system design across the organisation and its members.

Skills

Degree in CS/Math

Performance models

GPU/accelerator optimization

LLMs and DL hardware understanding

Profiling tools (Nsight, PyTorch Prof

Python for data (Pandas/NumPy)

Education

Degree in computer science, mathematics, or adjacent

Tools

CUDA

Nsight

PyTorch Profiler

Prometheus

Grafana

Job description

Pure Resourcing Solutions is seeking a Senior Performance Engineer to join a Cambridge-based AI infrastructure initiative. The role is hybrid, positioned between research and engineering to convert live training metrics into practical optimization insights.

You will create models and calculators forecasting performance, assess accelerator value, and guide architecture decisions that influence procurement and deployment across the organisation and its members.