Agentic Engineering Platform Engineer (n8n GmbH)
Darbdavys: n8n GmbH
Miestas: Vokietija
Paskelbta: 2026-10-21
💶 91400 - 171700 €/m. Prieš mokesčius · visa darbo diena · nuotolinis darbas: taip
Kandidatuoti šaltinyje (cv.lt) →
Job description:
Your main goal will be to keep n8n operating as an AI-native engineering organization where agents are embedded across the development cycle; helping teams plan, build, test, review, learn, and improve.
You’ll do this by maintaining and evolving the platform foundations, tools, guardrails, and team processes that make agentic development safe, measurable, and sustainable. Working with Developer Platform, Security, AI/Agents, Architecture, and product engineering teams, you’ll help teams continuously adopt what works from frontier AI and make success visible through faster feedback loops, better quality, stronger resilience, and more learning from what we ship:
Define safe task categories for autonomous agents, including what they can attempt, what they should never touch, and where additional review is required.
Design and implement isolated execution environments where agents can work without exposing secrets or accessing sensitive systems unnecessarily.
Build pilot workflows that allow selected Linear issues to be picked up by agents, tested, and converted into useful draft PRs.
Create and maintain repo instructions, agent playbooks, ownership metadata, coding guidelines, and documentation that improve agent performance.
Continuously improve developer workflows, repo structure, testability, and documentation based on where agents succeed, fail, or create reviewer friction.
Partner with Architecture and senior engineers to define boundaries, dependencies, and areas where agents need more context or tighter constraints.
Integrate agentic workflows with Linear, GitHub, CI, test infrastructure, and developer review processes.
Design PR evidence templates that help reviewers understand what changed, why, what tests ran, what risks remain, and where human judgment is needed.
Partner with Security, AI/Agents, Developer Platform, and product engineering teams to define guardrails for sensitive areas like auth, billing, permissions, and data handling.
Create benchmarks from real historical engineering tasks to evaluate agent output quality, failure modes, test pass rates, review burden, and production risk.
Define quality signals and adoption metrics such as draft PR usefulness rate, reviewer acceptance rate, CI pass rate, rework rate, and developer satisfaction.
Use data and feedback from pilot teams to improve agent workflows, reduce unsafe attempts, and build trust across engineering.
Requirements:
You have experience with autonomous coding agents, AI coding platforms, agent orchestration systems, or internal AI developer tooling.
You have practical experience building, deploying, or operating AI-assisted or agentic software development workflows that work with developers not besides them.
You have a strong software engineering background, ideally in platform engineering, developer experience, infrastructure, tooling, CI/CD, or engineering productivity.
You’re comfortable leading a first-of-a-kind technical initiative from problem…
Your main goal will be to keep n8n operating as an AI-native engineering organization where agents are embedded across the development cycle; helping teams plan, build, test, review, learn, and improve.
You’ll do this by maintaining and evolving the platform foundations, tools, guardrails, and team processes that make agentic development safe, measurable, and sustainable. Working with Developer Platform, Security, AI/Agents, Architecture, and product engineering teams, you’ll help teams continuously adopt what works from frontier AI and make success visible through faster feedback loops, better quality, stronger resilience, and more learning from what we ship:
Define safe task categories for autonomous agents, including what they can attempt, what they should never touch, and where additional review is required.
Design and implement isolated execution environments where agents can work without exposing secrets or accessing sensitive systems unnecessarily.
Build pilot workflows that allow selected Linear issues to be picked up by agents, tested, and converted into useful draft PRs.
Create and maintain repo instructions, agent playbooks, ownership metadata, coding guidelines, and documentation that improve agent performance.
Continuously improve developer workflows, repo structure, testability, and documentation based on where agents succeed, fail, or create reviewer friction.
Partner with Architecture and senior engineers to define boundaries, dependencies, and areas where agents need more context or tighter constraints.
Integrate agentic workflows with Linear, GitHub, CI, test infrastructure, and developer review processes.
Design PR evidence templates that help reviewers understand what changed, why, what tests ran, what risks remain, and where human judgment is needed.
Partner with Security, AI/Agents, Developer Platform, and product engineering teams to define guardrails for sensitive areas like auth, billing, permissions, and data handling.
Create benchmarks from real historical engineering tasks to evaluate agent output quality, failure modes, test pass rates, review burden, and production risk.
Define quality signals and adoption metrics such as draft PR usefulness rate, reviewer acceptance rate, CI pass rate, rework rate, and developer satisfaction.
Use data and feedback from pilot teams to improve agent workflows, reduce unsafe attempts, and build trust across engineering.
Requirements:
You have experience with autonomous coding agents, AI coding platforms, agent orchestration systems, or internal AI developer tooling.
You have practical experience building, deploying, or operating AI-assisted or agentic software development workflows that work with developers not besides them.
You have a strong software engineering background, ideally in platform engineering, developer experience, infrastructure, tooling, CI/CD, or engineering productivity.
You’re comfortable leading a first-of-a-kind technical initiative from problem…