Junior AI Infrastructure Performance Engineer – Hybrid Cambridge

Pure Resourcing Solutions; Cambridgeshire and Peterborough; Hybrid; GBP 63,000 - 77,000; Full time

Benefits offered by this job

Pension

Hybrid Cambridge office

Academic/AI community exposure

Job summary

Pure Resourcing Solutions seeks a Junior Performance Engineer to help build predictive models for AI infrastructure in Cambridge, collaborating with senior engineers on live training and inference workloads.

You will turn metrics into practical tools, assess whether optimizations are worth shipping, and own early-stage performance analyses with support. Hybrid Cambridge role with strong growth potential.

Qualifications

• Postgraduate research background, PhD preferred or exceptional master’s graduates considered.
• Strong understanding of computer architecture and how LLMs run on hardware.
• Experience building performance models or forecasting tools using Python or spreadsheets.
• Hands-on with GPU/accelerator code (CUDA or similar).
• Familiar with profiling tools (Nsight, PyTorch Profiler) and monitoring stacks (Prometheus, Grafana).
• Strong Python for data work; Pandas and NumPy; scripting ability.

Responsibilities

• Collaborate with senior engineers to pull metrics from live training and inference jobs.
• Translate metrics into models and calculators to assess whether an optimization is worth shipping.
• Own early-stage performance assessments with close senior support.
• Evaluate cost-effectiveness of different setups before deployment.

Skills

PhD preferred

Computer architecture

Performance modeling

CUDA / GPU

Profiling tools

Python + Pandas

Education

PhD in CS/Math/Physics

Master degree in related field

Tools

Nsight

PyTorch Profiler

Prometheus

Grafana

CUDA

Job description

Pure Resourcing Solutions seeks a Junior Performance Engineer to help build predictive models for AI infrastructure in Cambridge, collaborating with senior engineers on live training and inference workloads.

You will turn metrics into practical tools, assess whether optimizations are worth shipping, and own early-stage performance analyses with support. Hybrid Cambridge role with strong growth potential.