DevOps Engineer

• Asks “how do I make this impossible to get wrong” rather than “how do I fix this instance”
• Takes proactive initiative on environment health and automation without waiting for direction
• Gets satisfaction from other people shipping safely at speed, and builds controls that unblock rather than obstruct
• Uses AI aggressively while assuming its output is wrong until verified, and encodes that assumption into the checks
• Explains technical issues clearly to non-technical colleagues and, where needed, client contacts
• Stays analytical during high-severity production incidents

• Hands-on experience managing and configuring AWS environments - ECS (Fargate), RDS, VPC, S3, API Gateway, Lambda, CloudFront, IAM
• Hands-on Infrastructure-as-Code experience with Terraform or OpenTofu, including established module patterns, state management, and safe handling of destructive changes
• Containerisation experience with Docker, including image optimisation and secure runtime configuration
• Demonstrated experience building and maintaining CI/CD pipelines and deployment automation, including automated testing and rollback
• Strong command-line, scripting and programming ability - Linux, Bash, Git, plus Python, TypeScript or JavaScript - for diagnostics and automation
• Solid grounding in information security best practice, secrets management, system documentation, and operating resilient systems with no single points of failure
• Commercial experience supporting multiple environments under SLA and support commitments
• Practical, current experience using AI coding agents (Claude Code or equivalent) as the primary means of producing code, and the judgement to verify their output rather than trust it

• Team: Product Engineering - DevOps. Five engineers including this role, reporting to the Senior DevOps Architect
• On-call: Out-of-hours rota as first responder, currently one week in six

This is a hands-on DevOps role on our Product Engineering team, and it’s built around two things.

First - Alliants engineering is AI-assisted by default, and the function is shifting from doing the work directly to building the controls that let the wider organisation ship safely. You’ll build and maintain the automated validation, testing, standards enforcement and risk detection that stand between generated code and production. Your work decides what everyone else can safely build.

Second - Ownership. You’ll deploy, maintain and optimise the AWS infrastructure behind our proprietary products and client solutions and hold it to 99.9%+ availability, security compliance, and no single points of failure from network through to database.

Role Focus & Operating Context All code in this role is produced with AI coding agents. A thorough understanding of infrastructure-as-code and automation is essential. The skill we hire for is directing, constraining, reviewing and verifying output to a high standard.

Standards, architectural decisions and runbooks are treated as machine-readable inputs to agents, not just documents for humans. They have to stay accurate and enforceable as the volume of change keeps rising: every control must scale without a human in the loop for the routine case, while reliably stopping the dangerous one. You’ll work across proprietary products and client hosting environments, under real SLA and support commitments, alongside four other DevOps engineers and the software engineering teams who consume what you build.

What you’ll work on and own

• The guardrails - policy-as-code, protected environments, scoped IAM, blast-radius limits and promotion paths that make the dangerous change impossible rather than discouraged
• The automated quality layer - static analysis, schema and migration safety checks, dependency and licence scanning, automated code review and integration testing
• The infrastructure - AWS environments defined in OpenTofu/Terraform, for Alliants products and for clients
• The pipelines - fast, parallel, self-verifying CI/CD with automatic failure detection and rollback to last known-good
• The environments assigned to you - availability, security posture, cost, and what happens at 3am when something breaks
• The templates and standards that make the safe route the easiest route for everyone else