Agentic Tooling Engineer EU/UK remote (m/f/d) remote

Description

The company is a modular payments platform for businesses. It combines commercial cards, payment accounts, FX and spend management in one system that works with existing setups.

Customers in travel, e-commerce and insurance run card-based payment workflows on it, backed by credit lines the company underwrites itself.

Fintechs and software platforms build on its payment infrastructure, from API-based card services to fully white-label products. Banks run its customer-facing spend interface on their own systems.

Founded in 2020 and based in Berlin, the company serves more than 5,000 businesses and 40+ partners across Europe and the US. It is a principal member of Visa and Mastercard, issues commercial cards in 13 currencies across 32 countries, and holds its own European e-money license plus licensed partners in the UK and US.

The tech organisation has more than 100 people and is moving to one shared agentic framework for taking work from idea to production with AI agents, with clear human decisions and automated checks.

You will join the CTO office, reporting to the CTO and working alongside Developer Experience. You will own the shared infrastructure all agentic tooling depends on and help make the framework reliable, measurable and easy to adopt. This is not a prompt engineering job: the framework is an internal product with versions, evals and users.

Responsibilities

• Build and maintain the framework's shared tooling: distribution and installation across agent tools, shared skills and agents, lints, eval runners, and the interfaces between harnesses.
• Help teams building their own harnesses, skills and agents follow the framework's requirements, and carry good ideas from one tool to another.
• Create evals and certification so a team can show that a harness change is better before it ships.
• Together with Developer Experience and the other Platform Infrastructure teams, move stable rule-based steps out of prompts and into platform checks that run for every author and every tool.
• Collect friction from real use, fix its cause, and add a test so the failure stays fixed.
• Drive adoption across the tech organisation by pairing with teams, training people and measuring delivery outcomes.
• Suggest and build framework improvements, and help define new processes and agents on top of it.

Requirements

• 7+ years of software engineering on production systems, including developer tooling or an internal platform with real users. At Staff level, you have also set technical direction across several teams.
• Deep, daily hands-on use of a wide range of agentic tools, such as Claude Code, Codex, Cursor, Windsurf, Aider, Devin and CodeRabbit, and of the infrastructure agents run on. You know where each is strong and where it fails.
• Agentic tooling you built that other engineers use, such as skills, plugins, hooks, subagents, MCP servers or custom agents.
• Experience with LLM evals and context engineering: evals for non-deterministic systems, reading failure traces, and how the size, freshness and consistency of context change agent behaviour.
• Strong system design and a verification mindset, knowing which controls must be deterministic and which can rely on a model.
• Strong Python, git and GitHub (Actions, rulesets, CODEOWNERS). Reading Java and Kotlin comes easily.
• Clear technical writing and the ability to change how other engineers work without formal authority over them.
• A plus: experience in a regulated environment such as payments or banking, a security engineering background, or an open-source project you maintained.

Conditions

Remote role open to candidates in the EU or UK, with flexible remote working.

Equal opportunity employer: all backgrounds, identities and abilities are welcome, and interview accommodations are available on request.

Resumes are reviewed against the role's qualifications using Ashby AI criteria, but the Talent and Hiring Teams make every final decision and Ashby never rejects anyone automatically.

Benefits

• Growing team with large responsibilities, a strong culture of knowledge sharing and high standards.
• Attractive pay.
• Choice of Windows or Mac.
• Flat hierarchy and open communication in a relaxed, professional setting.
• Room to grow in a dynamic team with ambitious goals.
• Monthly mobility benefit.
• Wellhub membership.
• Company card with a monthly credit to try the product and buy food with colleagues.