AI Infrastructure Validation Engineer for GPU Clusters

Carbon3.ai; Greater London; On-site; GBP 90,000 - 120,000; Full time

Job summary

Era4 is seeking an AI Infrastructure Validation Engineer to design, build, and orchestrate automated preflight validation suites and performance benchmarks for bare-metal APIs, Kubernetes, and multi-node GPU clusters. You will ensure every platform release and hardware deployment is tested and production-ready before reaching customers.

You will collaborate across Product, Engineering and Operations, building resilient test orchestration, observability pipelines, and integrated validation within

Qualifications

• Strong proficiency in Python for validation tooling.
• Hands-on experience with Infrastructure as Code to provision bare-metal and containerized environments.
• Proven experience with Kubernetes and enterprise-scale distributed systems.
• Experience integrating validation suites into CI/CD pipelines.
• Experience with observability and real-time diagnostics.

Responsibilities

• Design and build Python-based preflight verification tools for multi-node, GPU-enabled workloads.
• Write and maintain IaC to provision and test complex bare-metal and containerized compute environments.
• Create adaptive test orchestration to detect environment drifts and analyze platform changes.
• Automate cluster benchmarking and low-latency network validation for node-to-node throughput.
• Integrate validation into CI/CD pipelines and maintain observability pipelines for logs and metrics.
• Collaborate with Platform, Network, and Datacentre Operations to root-cause platform regressions.

Skills

Python

Infrastructure as Code

Kubernetes

CI/CD

Observability

Root-cause analysis

Tools

Redfish

BMC

IPMI

Job description

Era4 is seeking an AI Infrastructure Validation Engineer to design, build, and orchestrate automated preflight validation suites and performance benchmarks for bare-metal APIs, Kubernetes, and multi-node GPU clusters. You will ensure every platform release and hardware deployment is tested and production-ready before reaching customers.

You will collaborate across Product, Engineering and Operations, building resilient test orchestration, observability pipelines, and integrated validation within