Performance Engineer (Junior) | AI Infrastructure | Cambridge (Hybrid) |

Overview

In this junior role you will work with senior engineers to convert live training and inference metrics into predictive models and decision tools. You’ll help determine whether alternative hardware setups and configurations are worth shipping, contributing to cost-aware, performance-driven AI infrastructure. The position offers hands-on ownership early and close mentorship, within a Cambridge-based, research-connected organization.

Pay / Benefits

• Pension
• hybrid from Cambridge office
• exposure to wider AI and academic scene

Responsibilities

• Collect real metrics from live training and inference jobs
• Build performance models and calculators to answer questions about deployment viability and cost
• Collaborate with senior engineers and receive close guidance
• Own small-to-medium scope modeling tasks and iterate towards tangible predictions

Key requirements

• Postgraduate research background; PhD preferred, master’s considered
• Strong grasp of computer architecture fundamentals and how LLMs and DL models run on hardware
• Experience building performance models or forecasting tools (Python or spreadsheets)
• Hands-on CUDA or similar GPU/accelerator code experience
• Familiarity with profiling tools (Nsight, PyTorch Profiler) and monitoring stacks (Prometheus, Grafana)
• Strong Python for data work; Pandas and NumPy
• Exposure to inference serving frameworks like vLLM is a plus

• strong problem-solving abilities
• quick learner
• collaborative and proactive communication
• CUDA
• Nsight
• PyTorch Profiler

…