Senior Internal GenAI engineer

Level

Mid (2–4 years relevant experience)

ABOUT THE TEAM

The GenAI Enablement Team is the centre of excellence for generative AI inside the organisation. We build, evaluate, and roll out the AI tools our colleagues use every day — from M365 Copilot and Google Gemini Enterprise to Claude Desktop, ChatGPT Enterprise, and emerging tools like Codex, Cowork, and Claude Code. We also build the custom MCP servers that connect those tools to our internal systems, and publish the guidance and patterns that let teams across the business get value from them safely.

ABOUT THE ROLE

You’ll build and ship custom MCP servers independently, run structured evaluations of new AI tooling, and produce the skills, prompts, and documentation that help colleagues across the business get real value from their AI tools. This is a hands-on engineering role with a strong enablement flavour — you’re expected to own features end-to-end, proactively raise quality issues, and contribute to the patterns the wider team builds against. You’ll work closely with the Lead GenAI Engineer, taking on meaningful parts of larger projects with a genuine degree of autonomy.

WHAT YOU’LL DO

·       Build and maintain custom MCP servers in Node.js (TypeScript) or Python, working independently — connecting our SaaS systems (Confluence, Jira, Salesforce, ServiceNow and others) to Claude Desktop, ChatGPT Enterprise, Cowork, and Codex — taking full ownership of auth, error handling, and observability.

·       Run structured evaluations of new releases across our AI tool stack — designing test cases, scoring outputs against defined rubrics, and producing clear written recommendations for the team.

·       Write skills, prompts, and documentation that turn raw tool capability into reusable patterns for non-technical colleagues; own specific documentation areas from first draft through to ongoing maintenance.

·       Contribute to Copilot Studio agents and Microsoft Foundry pipelines, exercising sound architectural judgement between MCP-first and Copilot-first approaches depending on the use case.

·       Support security and platform reviews for new tooling — assist with DPIA documentation, data-residency checks, and access-control reviews alongside senior engineers.

·       Own a share of #ai-help triage, systematically converting recurring questions into FAQ entries, skill updates, or targeted MCP improvements rather than answering the same question twice.

·       Contribute a developer’s perspective to product reviews and feature enablement decisions as our AI suite evolves — helping the team stay grounded in what’s practical to build, maintain, and scale.

WHAT WE’RE LOOKING FOR

Essential

·       2–4 years building production software in Python and/or Node.js (TypeScript or JavaScript).

·       At least one demonstrable AI integration you’ve shipped — an MCP server, Copilot Studio agent, custom GPT, Claude skill, or comparable LLM tool that real users have relied on.

·       Working familiarity with the tools we operate — you can articulate the differences between Claude Desktop’s skill model, Copilot Studio’s plugin model, and the Codex / agentic-coding surface, and have used several in anger.

·       Comfortable owning a feature end-to-end — scoping, building, testing, documenting, and shipping — with a track record of doing so without needing to be chased.

·       Confident on a developer’s command line: git, npm/pnpm or pip/uv, REST/HTTP debugging, CI/CD basics, and reading error logs without guidance.

·       Clear written communicator — your PRs, design notes, and documentation pages are easy to follow by someone who wasn’t in the room.

Nice to have

·       Familiarity with at least one cloud platform (Azure preferred) and identity / auth patterns (OAuth 2.0, Entra ID, Okta, SSO).

·       Experience with structured eval tooling — promptfoo, RAGAS, LangSmith, Braintrust, or a comparable harness.

·       Light data skills — pandas, SQL, or basic analytics to support evaluation work.

·       Exposure to regulated-industry constraints such as data residency, model-output governance, or DPIA processes.

·       Public-facing artefacts (GitHub projects, blog posts, conference talks) that demonstrate your thinking about AI tooling.

U.S. National Base Pay Range: $86,600 - $144,400. Geographic differentials may apply in some locations to better reflect local market rates. This job is eligible for an annual incentive bonus.

We know your well-being and happiness are key to a long and successful career. We are delighted to offer country specific benefits. Click here to access benefits specific to your location.

We are committed to providing a fair and accessible hiring process. If you have a disability or other need that requires accommodation or adjustment, please let us know by completing our Applicant Request Support Form or please contact 1-855-833-5120.

Criminals may pose as recruiters asking for money or personal information. We never request money or banking details from job applicants. Learn more about spotting and avoiding scams here.

Please read our Candidate Privacy Policy.

We are an equal opportunity employer: qualified applicants are considered for and treated during employment without regard to race, color, creed, religion, sex, national origin, citizenship status, disability status, protected veteran status, age, marital status, sexual orientation, gender identity, genetic information, or any other characteristic protected by law.

USA Job Seekers:

EEO Know Your Rights.