Staff LLM Interaction Engineer

The AI orchestration of your wildest imagination. n8n is the open workflow orchestration platform built for the new era of AI. We give technical teams the freedom of code with the speed of no-code, so they can automate faster, smarter, and without limits. Backed by a fiercely inventive community and 500+ builder-approved integrations, we’re changing the way people bring systems together and scale ideas for impact. Since our founding in 2019, we’ve grown into a diverse team of over 160 - working across Europe and the US, connected by a shared builder spirit and with our centre of gravity in Berlin. Along the way, we’ve: Cultivated a community of more than 650,000 active developers and builders Earned 145k+ GitHub stars, making us one of the world’s Top 40 most popular projects Been ranked as one of Europe’s most promising privately held SaaS startups (4th in Sifted’s 2025 B2B SaaS Rising 100) Raised $240m to date, from Sequoia’s first German seed to our recent $180m Series C - bringing us to a $2.5bn valuation And are grateful for our 94 eNPS score (most companies would call 70 excellent) That’s the company we’ve built. Now we’d love to see what you can build. If you’re applying, try n8n out - whether you’re technical or not - and share a screenshot of your first workflow with us. The easiest place to start is here: app.n8n.cloud/register.We’re in a defining moment of an incredible journey. Come and build with us.Role As a Staff level LLM Interaction Engineer, you’ll help shape how n8n’s AI Workflow Builder works end-to-end. This staff level engineering role is focused on a critical product challenge: allowing both technical or less-technically proficient users of n8n to turn AI powered automated workflows into something they can rely on every day. That means building systems that consistently generate the right output, stay grounded in the user’s workflow, and improve quickly through iteration and evaluation. You’ll work on the orchestration layer between users, workflows, and LLMs—designing the context, tool interactions, and evaluation loops that make the product feel fast, helpful, and trustworthy. What you’ll do Architect and implement AI-assisted workflow building capabilities (workflow creation, editing, and suggestions) Design systems that provide LLMs with the right context: workflow state, user intent, constraints, and history Build tool-driven agent behavior: calling internal tools, validating outputs, correcting mistakes, and recovering gracefully Develop evaluation benchmarks and automated testing for AI output quality Create fast iteration loops: experiment, measure, improve—shipping changes quickly with confidence Collaborate closely with engineering, product, and design to deliver a polished Copilot experience Stay current with advances in LLMs and AI developer tools and bring practical improvements into the product Requirements 7+ years of experience building web-based products, ideally SaaS developer tooling products Strong fullstack engineering skills in TypeScript / Node.js and API design Hands‑on experience building complex AI-powered product features in production Experience building a tool‑using AI system or “agentic” workflow, such as:coding agents, AI copilots, or assistant systems multi‑step tool calling pipelines structured generation with validation and recovery Strong understanding of how to make LLM outputs reliable:prompt design and iteration context construction and optimization managing constraints, failure modes, and edge cases Experience building or using evaluation frameworks for LLM‑based systems:benchmarks, scoring, regression tests, offline vs online evaluation Solid understanding of what to evaluate A product mindset: you care deeply about user outcomes, not just model behavior A bias for action: you iterate quickly, learn fast, and ship improvements continuously Bonus points for Experience with developer tools that generate code or structured output (workflows, config, SQL, DSLs) Familiarity with agent orchestration patterns, state machines, or DAG‑based systems Experience with AI reliability tooling: trace/debug pipelines, test harnesses, and observability for LLM systems Contributions to open‑source AI tooling (evals, orchestration, prompt management) Experience with RAG systems and retrieval patterns (when helpful) Benefits Competitive compensation – We offer fair and attractive pay. Ownership – Our core value is “empower others,” and we mean it—you’ll get a slice of n8n with equity. Work/life balance ️ – We work hard but ensure you have time to recharge:Europe: 30 days of vacation, plus public holidays wherever you are. US: 15 vacation days, 8 sick days, plus public holidays wherever you are. Health & wellness