Cluster Automation Engineer for Reproducible AI (Hybrid)

CommonAI CIC; Cambridge; Hybrid; GBP 65,000 - 90,000; Full time

Benefits offered by this job

Competitive salary

Pension

Professional development

Hybrid work arrangement

Office near Cambridge train station

Job summary

CommonAI CIC is seeking engineers to build software that configures and controls real AI infrastructure for reproducible experiments. You will expose cluster controls through programmatic interfaces and automate configuration and experiment setup across the stack.

Work with measurement and simulation engineers and cluster operators to implement reliable execution with state checks, error handling and recovery, enabling repeatable experiments and meaningful results.

Qualifications

• Experience building systems software or automation that configures and controls real computing infrastructure.
• Strong programming and debugging skills with practical understanding of Linux and hardware/software interfaces.
• Depth in cluster/HPC automation, distributed systems, orchestration, device control, networking, or experiment-control software.

Responsibilities

• Identify and expose controls on the cluster through documented programmatic interfaces.
• Automate configuration and experiment setup across the stack (workload placement, serving settings, resources, device config).
• Build reliable execution with precondition checks, state confirmation, error handling and recovery to a known configuration.
• Record actions, outcomes and configurations for reproducibility of experiments.
• Coordinate access and changes with cluster operators and test success/failure/recovery paths.
• Design interfaces for future optimisation software with clear limits and outcomes when an action cannot be completed.

Skills

Systems software

Automation

Linux

Kubernetes

Slurm

Terraform

AI coding agents

Distributed systems

Debugging

Tools

NixOS

CoreOS

OpenTofu

Terraform

Job description

CommonAI CIC is seeking engineers to build software that configures and controls real AI infrastructure for reproducible experiments. You will expose cluster controls through programmatic interfaces and automate configuration and experiment setup across the stack.

Work with measurement and simulation engineers and cluster operators to implement reliable execution with state checks, error handling and recovery, enabling repeatable experiments and meaningful results.