Hybrid AI Platform Engineer: Inference & GPU Infra

Leonardo UK Ltd; Easter Howgate; Hybrid; GBP 70,000 - 110,000; Full time

Benefits offered by this job

Flexi-days

Pension scheme

Mental health support

Referral bonus

Online courses

Job summary

Leonardo UK Ltd is recruiting AI Platform Engineers across all levels to shape, build and operate the infrastructure underpinning our AI platform in the UK.

You will own the platform layer above GPU infrastructure, deploying scalable inference services with containers and Kubernetes, while ensuring security, performance and reliability. Hybrid working is available across Edinburgh and other UK sites.

Qualifications

• Experience operating shared or business-critical platforms.
• Familiarity with large language models and running them.
• Experience troubleshooting distributed systems and performance issues.
• Familiarity with infrastructure as code and CI/CD.
• Effective communication and stakeholder engagement.
• Curiosity and a drive for continuous improvement.

Responsibilities

• Take ownership of the platform layer above our managed GPU infrastructure, including model serving, inference runtimes, AI gateways and supporting services.
• Using technologies such as containers, Kubernetes, vLLM and AI gateway platforms, deploy and operate scalable inference services, improve observability and performance, and investigate complex technical issues across the platform.

Skills

Platform operation

LLMs familiarity

Distributed systems

IaC CI/CD

Stakeholder communication

Curiosity and improvement

Tools

Docker

Kubernetes

vLLM

LiteLLM

Bifrost

SGLang

Job description

Leonardo UK Ltd is recruiting AI Platform Engineers across all levels to shape, build and operate the infrastructure underpinning our AI platform in the UK.

You will own the platform layer above GPU infrastructure, deploying scalable inference services with containers and Kubernetes, while ensuring security, performance and reliability. Hybrid working is available across Edinburgh and other UK sites.