Original listing text, shown exactly as published by the company.
What you'll do
Strategy & Prioritisation
- Own your function's AI strategy, aligned with your Executive Sponsor and business priorities. Understand which metrics matter to the org, identify what will move the needle, define how you'll measure impact, and track progress over time.
- Map how work flows across your function end-to-end, including the handoffs upstream and downstream to other orgs. Identify where the real constraints are, not just the ones your team can see. Focus on the 100x problems: where could leveraging AI in a workflow let your org execute it orders of magnitude faster, or at 100x more volume than before?
- Manage intake of AI requests, ideas, and pain points from across the function, including via your Champion network. Ensure every team member has a clear route to surface what they need, rather than building independently.
- Prioritise strategically against business outcomes and executive guidance. Hold the line on priorities - we cannot change direction every two weeks - and ensure the AI Engineer's time is spent on the highest-impact work.
Adoption & Change Management
- Reimagine, not just automate. Challenge your org to think beyond injecting AI into existing workflows. Work with Enterprise AI to spot opportunities to fundamentally rethink how work gets done.
- Drive adoption and change management together with the AI Engineer. The best AI solution is worthless if nobody uses it. Create the channels, rituals, and feedback loops that make AI visible in your function: shared spaces for teams to show what they've built, regular office hours, onboarding for new hires, and celebration of wins. Own the rollout and iteration needed to make AI initiatives stick.
- Coordinate with Enterprise AI to ensure your function benefits from patterns, tools, and learnings emerging across other parts of the business.
- Build and bridge the Champion network in your function. Champions are the peer community that extends your reach beyond what you and the AI Engineer can deliver directly. From early in the role, identify and recruit Champions across sub-teams (5-10% time, formally agreed with their manager), run a regular Champion sync, host demos to the wider function, and act as their bridge to Enterprise AI. Champions are not your reports: you coordinate them, you don't manage them. Without this network, your reach is capped.
Hands-On Building & Agent Operations
- Build AI agents using no-code and low-code platforms (e.g. Glean, Workato, similar tools). Go from idea to working prototype without waiting for engineering.
- Author and iterate on skills files that define how AI agents behave. Refine instructions based on real usage and share reusable skills across the function.
- Configure MCP servers and tools, giving agents access to the business systems they need. Partner with the AI Engineer on what to connect and how to do it securely.
- Own your function's fleet of agents. Some agents will be used directly by people in your org. Those that aren't, you own. Either way, you're accountable for their performance: tracking KPIs, running evaluations after model or data changes, and iterating based on what you learn.
- Expect to rebuild. AI tools and models evolve fast. The agent you built last month may need to be replaced, not patched. You should be comfortable sunsetting your own work when a better approach emerges, and helping your org stay current rather than attached to what exists today.
What you'll bring
Business & Strategic
- Deep knowledge of your function's operations. You understand how work moves through your org, where it gets stuck, and why. You can trace a process end-to-end and explain how it connects to the teams around it.
- Strategic prioritisation skills. You've managed competing demands before and can make hard calls about what matters most. You think in terms of business outcomes, not activity.
- A product management mindset. You naturally think about intake, backlog, iteration, and adoption. You're comfortable defining success metrics and holding yourself accountable to them.
- Strong communication and influence. You'll be the person saying "not yet" to some teams and "think bigger" to others. You need the credibility and interpersonal skills to make both of those conversations land.
- Cross-functional instincts. You default to understanding how your org's processes affect and are affected by the teams around you - not just optimising in isolation.
- Experience building peer networks or communities of practice. You've recruited and sustained volunteer contributors before, whether as a guild lead, champion programme owner, or community organiser. You know how to motivate people whose time you don't directly own.
Hands-On & Technical
- Comfortable building with AI tools. You don't need to write production code, but you should be able to build a working agent, configure a skill, connect an MCP server, and troubleshoot when something isn't working. Think: power user, not software engineer.
- Ready to learn fast. Experience with or willingness to quickly pick up no-code/low-code AI platforms, prompt engineering, and agent configuration. You'll be trained on GitLab's specific tooling, but you should arrive ready to get your hands dirty.
- Strong conceptual understanding of AI capabilities - summarisation, classification, generation, automation, agentic workflows - and a commitment to staying current. The landscape shifts constantly. The tools you use today may be obsolete in weeks. You stay on top of new developments so that your org's AI strategy reflects what's actually possible, not what was possible six months ago.
- Ability to map data flows - structured and unstructured. Understand where agents need context, and figure out where humans should interface with automated workflows and at what steps.
