Confirmed on the employer's own hiring board on Nov 18, 2026. First seen by Alion on Nov 18, 2026.
Job Introduction
Senior Cloud Platform Engineer
Location: UK Remote with visits to our Sheffield HQ, 6 times a year
Contract: Permanent
Hours: 37.5 hours (Core hours 10:00 am-2:00 pm Monday-Friday)
Trusted by teaching communities worldwide, Twinkl provides instant access to a complete range of teaching, planning and assessment materials that support learning from birth through to secondary education and beyond.
Our tech organisation is now maturing how it builds and runs digital products, and AWS is at the centre of that.
Cloud Platform Engineering is a small team that owns the cloud foundations every Twinkl engineering team builds on. We run the AWS organisation, the shared Terraform modules and CI/CD pipelines that product teams deploy with, and the edge in front of Twinkl's sites: WAF, CDN and DNS. We also look after observability and on-call, security and disaster recovery, and cloud cost. Everything is managed as code and changed through pull requests.
We provide a platform for the teams building in AWS, and we work on a "you build it, you run it" principle. Product teams own and operate their own services in production. Our job is to give them the platform, guardrails and paved paths that make that safe and straightforward, not to run their services for them.
That makes us an enabling team as much as a build team. Product teams come to us for support and guidance, and we lead the rollout of AI developer tooling across engineering.
As a senior engineer you will own pieces of this platform end to end. You will turn business requirements into technical designs, favour simple solutions to complex problems, and mentor the engineers around you.
What you'll do:
You will do all of this with Claude doing the heavy lifting. We work agent-first: you set the direction, make the judgement calls and own the outcome, while Claude does the bulk of the writing, investigating and checking. That leaves you free to focus on innovation and on delivering value to our customers.
Build for agentic developers first. We assume the engineer using the platform is working alongside an AI coding agent, and we design for that from the start. You will make our modules, pipelines and standards something an agent can discover, use correctly and be safely stopped by when it gets things wrong: skills and connectors that encode how Twinkl builds, automated review and policy checks in CI, least-privilege access for agents, and telemetry that shows what is working.
Run and evolve the AWS organisation. A multi-account estate with account provisioning, identity and just-in-time access, organisation-wide guardrails, and keyless federation from CI into AWS. Every change is Terraform, reviewed in a pull request, and applied by a pipeline.
Build the paved path, and get it adopted. A versioned library of Terraform modules, reusable GitHub Actions and deployment pipelines that let product teams ship infrastructure and applications on their own. The harder half is adoption: migrating existing services onto the modules, measuring coverage, and retiring the one-off patterns they replace.
Own the edge. WAF, CDN and DNS across Twinkl's international domains. That means telling scrapers and bots from real teachers and legitimate crawlers, diagnosing blocked requests from evidence in the logs, and tuning protection without hurting customers or search traffic.
Make production observable. Metrics, logs, traces and alert routing that product teams use to run their own services, in line with "you build it, you run it". You will take a share of the platform on-call rota, lead incidents when they are ours, and turn what we learn into fixes.
Raise security and resilience. Backup and recovery standards, multi-region disaster recovery and failover exercises, least-privilege IAM, and assisting teams with diagnosing security findings across every account.
Manage cloud cost. Attribute spend to teams and products, investigate anomalies, report to senior leadership, and shape commitment and right-sizing decisions with real usage data.
Enable other teams. Handle platform support requests, review designs, and mentor junior engineers. When the same request arrives a third time, turn it into self-service.
Essential Skills & Experience:
AWS at depth. You have run a multi-account AWS organisation in production, and are confident with identity and access, organisation-wide guardrails and networking.
Terraform as your default. You write reusable modules that other teams depend on, not only consume them, and you expect every change to go through a pull request and a pipeline.
CI/CD. You have built and maintained deployment pipelines, ideally in GitHub Actions, including secure keyless access from CI into the cloud.
Agent-first working. You already use an AI coding agent such as Claude Code for real engineering work. You know how to direct it, review what it produces and verify the result before it ships.
Operating production. You have set up monitoring and alerting, been on call, and led incidents through to the fix.
Security and resilience. Least-privilege access, backup and disaster recovery are things you design in, and you can help another team diagnose a finding in a system you did not build.
An enabling mindset. You are comfortable with "you build it, you run it": you would sooner give a team the tools and guidance to run their own service than run it for them. You write clearly and enjoy mentoring junior engineers.
Simple solutions. You can take a complex problem or a loose business requirement and find the simplest design that solves it.
Desirable Skills & Experience (Beneficial but not required):
Experience rolling out and supporting developer experience in an agentic environment: AI coding agents, the tooling and guardrails around them, and helping engineers get the most from them.
Experience at the edge: WAF, CDN and DNS, and telling bot traffic from real users.
Cloud cost management: attributing spend, investigating anomalies and making commitment decisions.
Serverless architecture on AWS.
Some software development experience.
How we hire:
Intro chat (30 minutes). A conversation with the Cloud Platform Engineering Manager, so you can find out more about the role.
Technical test (90 minutes). A practical session on the kind of problems the team works on.
Behavioural interview (90 minutes). Competency-based questions about how you work, with time for your questions about the role, the tech team and Twinkl.
We then review everything and let you know. Our team is on hand to answer questions at any point in the process.
If you require a reasonable adjustment to the application/selection process to enable you to demonstrate your ability to perform the job requirements please let us know. This will help us to understand any modifications we may need to make to support you throughout our selection process.
How We’ll Look After You:
Westfield Health Cash Plan: Allows users to claim money back (up to set limits) for everyday healthcare expenses, such as dental check-ups, optical care, and therapy treatments.
Annual Leave: 33 days inclusive of bank holidays (pro rata).
Me Day: A dedicated “Me Day” each year for yourself.
Volunteering: A paid charity volunteering day each year.
Twinkl Perks: Free Twinkl subscriptions for you and your family.
Culture: A friendly, welcoming, supportive environment and a people-first culture where collaboration and kindness matter.
Flexibility: Flexible working with opportunities to work from home.
Inclusion: A commitment to creating an inclusive workplace. Our Employee Network Programme includes groups focused on LGBTQ+, People of Colour, Disabilities (visible/invisible), Women in Tech & Working Parents.
Financial Security: Enhanced pension contributions and Life insurance.
Family Support: Enhanced maternity, adoption, and paternity pay.
Loyalty: Long service awards and additional annual leave.
Apply for this job

