Job Description

Who we are

We are looking for a hands-on AI Engineer to help improve and scale an internal platform that enables engineering teams to ship faster and safer using AI-powered workflows.

This role is focused on practical AI engineering, not AI research and not pure platform architecture. You will work on agentic workflows, prompt engineering, evaluation pipelines, GitHub-based automation, and developer tooling that helps teams adopt AI agents consistently across repositories.

You will join an existing initiative that already has working foundations in place. The immediate need is to strengthen, stabilise, and mature the current implementation through hands-on engineering, especially in agent development, testing, evaluation, and workflow reliability.

This is an excellent fit for someone who already uses AI coding agents as part of their daily workflow and knows how to turn AI capabilities into reliable engineering tooling.

What you'll be doing

• Build and improve AI workflows across repositories
• Design agent workflows for docs, testing, code quality, reviews, and security
• Create effective prompts and multi-step interactions
• Improve reliability and usability of agents
• Use AI coding tools (e.g., Copilot, Cursor) to accelerate delivery
• Design evaluation pipelines for prompts and agents
• Build datasets with edge and adversarial cases
• Add automated regression and behavior checks
• Use code-based and LLM-based evaluators
• Define quality gates before production
• Build reusable templates and onboarding tools
• Develop CLI/scripts for easy adoption
• Maintain workflow catalog and integrations
• Write clear docs and guides
• Enable self-service usage
• Maintain GitHub Actions and CI/CD patterns
• Ensure secure design (least privilege, secrets, OIDC)
• Improve security controls and automation safety
• Mitigate risks (prompt injection, data leaks)
• Keep changes testable and production-safe

What you'll bring along

• BSc/MSc in Computer Science or related field
• Minimum 3-6+ years as a Platform Engineer
• Strong hands-on experience with Python for scripting, automation, evaluation tooling, or integrations
• Practical experience building with LLMs, prompts, AI agents, or agentic workflows in real engineering or production contexts
• Daily use of AI coding assistants or coding agents such as Copilot, Cursor, Kiro, Windsurf, Claude Code, or similar
• Experience designing, testing, and iterating prompts for reliable task execution
• Experience building evaluation or testing approaches for non-deterministic AI outputs
• Strong experience with GitHub Actions or equivalent CI/CD tooling, including reusable workflows and pipeline-as-code patterns
• Experience working with YAML, Markdown, JSON, and configuration-driven systems
• Good understanding of authentication, authorisation, secrets handling, and least-privilege access patterns
• Ability to write clear technical documentation for developers
• Strong hands-on engineering mindset with the ability to improve existing implementations quickly and pragmatically
• Preferred / Nice to Have
• Experience with AI evaluation frameworks such as promptfoo, DeepEval, RAGAS, Azure AI Evaluation SDK, or similar
• Experience with agentic frameworks such as LangChain, AutoGen, CrewAI, OpenAI Assistants API, or similar
• Background in internal developer platforms, developer tooling, or developer experience engineering
• Experience with docs-as-code and static documentation sites
• Familiarity with infrastructure-as-code tools such as Terraform, Bicep, Pulumi, or CDK
• Understanding of software supply chain security, dependency scanning, action pinning, and SBOM concepts
• Awareness of AI safety concerns such as prompt injection, adversarial testing, and OWASP Top 10 for LLM