Junior Cloud & AI Engineer

About the role

We're a dynamic and innovative company running a modern technology estate — public-cloud infrastructure, self-hosted open-source platforms, custom-built business applications, and AI-powered tooling. As we grow, we're hiring a graduate or early-career engineer who wants to learn and build a career in a growing business.

This is an exceptional learning opportunity. You will see and contribute to every layer of a real production environment — from networking and Linux servers, through container orchestration and identity providers, to CI/CD pipelines and AI/LLM-integrated applications. You'll be mentored hands-on by an experienced engineering lead and develop genuine, transferable engineering skills that few graduate programmes give full visibility into.

Our mission for this hire: help build, automate, and harden the systems that the wider business depends on — with a focus on automation, cost efficiency, and resilient design.

You'll work across the full stack of modern infrastructure and applications. In a typical month you might:

Cloud infrastructure — operate and improve workloads on a major public cloud platform (compute, networking, managed databases, object storage, identity and access management); manage Linux servers running containerised workloads; configure reverse proxies, automate TLS certificates, set up VPNs.

Application platform — use code repositories with built-in CI/CD to deploy applications; operate a self-hosted platform-as-a-service that makes deployments routine; use container-management and workflow-automation tools to orchestrate environments.

Identity, secrets & security — operate a secret-management system that holds the credentials the business depends on; maintain a single sign-on identity provider that federates corporate identities to internal applications; configure cloud and SaaS app registrations for SSO and bot integrations.

Custom business applications — help operate and improve the low-code business applications that serve our paying customers; help operate the CRMs, content sites, and internal tools the wider team uses every day.

AI / LLM integration — use commercial AI APIs from inside developer environments to accelerate everyday work; help operate a self-hosted AI chat front-end that lets staff use AI privately; help build the AI-assistant capability we're integrating into our internal tooling — a chatbot that performs real actions against production applications, generates marketing content, and acts as a productivity layer for the team.

Documentation & operational discipline — source-controlled infrastructure (configuration in code, runbooks for incident recovery); disaster-recovery design and testing; maintenance windows and restore drills.

This is not a "fix the printer" role. From early on you will own real production work and see how business-critical decisions get made.

• Genuine curiosity about how things work. When you read about TLS, you want to know what the handshake is actually doing. When you use containers, you want to understand the kernel features they sit on. When you talk to an LLM, you want to understand attention and tokenisation, not just prompt-engineer.
• A bias to learning fast and shipping. You can pick up a new tool over a weekend if it solves a real problem. You re comfortable being wrong on Monday and right by Friday.
• Comfort with ambiguity. Real infrastructure is messy. You won t always have a perfect spec. You can read documentation, prototype, and confirm what works.
• Care about doing things properly. You appreciate that "I forgot to write it down" is what causes outages at 3 AM.

• Computer fundamentals: how an operating system works, what a process is, how memory is laid out, why disk I/O matters.
• Networking: TCP/IP, DNS, TLS, HTTP — not necessarily mastery, but enough to debug a "site isn t loading" problem methodically.
• Linux command line: comfortable in a shell. Can read a container build file, edit a config file, follow a system log.
• Git fluency — branches, merge requests, rebases.
• Curiosity about how Large Language Models actually work — not…