AI platform engineer

Zühlke is building one knowledge base for the whole firm: skills, project experiences and official documents that every colleague and every AI assistant can use. AI clients such as Claude Code or GitHub Copilot reach it through a single governed access layer. Behind that layer sit the knowledge sources and, from 2027, our business systems, each connected as an MCP server. As an AI Platform Engineer you build and run this platform.

• Build and operate the identity broker, the gateway and the MCP composer on Azure Kubernetes Service.
• Implement OAuth 2.0 and OIDC flows with Keycloak, federated to Entra ID, including token exchange (RFC 8693) and per-source permissions.
• Connect new knowledge sources and business systems as MCP servers and help their teams onboard on their own.
• Build semantic search and indexing on PostgreSQL with pgvector, and improve ranking over skills, experiences and documents.
• Secure the MCP layer against tool poisoning and toxic flows between sources, and control token cost with budgets and rate limits per client.
• Run everything as code: infrastructure in Bicep, deployments through GitLab CI and Argo CD, realm configuration and access policies in Git.
• Own production readiness together with security and operations: audit, monitoring, load tests and autoscaling.
• Reduce the single-operator risk by documenting, automating and sharing operations knowledge.

• At least five years of professional software engineering, including running production services on a cloud platform.
• Strong Java and Spring skills;
• Solid knowledge of OAuth 2.0 and OIDC, ideally with Keycloak and Entra ID.
• Hands-on experience with Kubernetes and Azure (AKS, PostgreSQL, Key Vault, Front Door, Container Registry).
• Infrastructure as code and GitOps experience, for example Bicep, GitLab CI and Argo CD.
• Experience with the Model Context Protocol (MCP) or a strong interest in building MCP servers, including their security risks such as tool poisoning.
• Experience with search and retrieval, for example embeddings, pgvector or similar vector search.
• Observability and testing practice: Application Insights or similar, LLM tracing with Langfuse or similar, load tests with k6 or similar.
• You use AI coding agents in your daily work and can judge where AI-generated code needs extra review, especially in the identity path.
• Working knowledge of responsible AI and AI governance, for example the EU AI Act obligations on logging and human oversight.
• A security-first mindset and the discipline to keep a platform simple.
• Good communication skills and the ability to work with source owners, security and operations.
• You communicate confidently in English and work effectively in an international environment.
• Будет плюсом: experience with Spring Cloud Gateway is a plus; Experience with token exchange is a strong plus; Familiarity with agent frameworks such as Claude Agent SDK, Azure AI Foundry or LangGraph, whose agents are the clients of the platform, is a plus

• Условий в вакансии нет