Agentic AI Architect (m/f/d)

Responsibilities:

• Architect the Agent Backbone: Define and maintain the enterprise reference architecture for autonomous agents, ensuring seamless interoperability between managed services (Azure AI Foundry) and custom open-source runtimes on AKS.
• Define Agentic Standards: Own the technical standards for the ecosystem, including the Model Context Protocol (MCP) for tool interoperability and the Agent-to-Agent (A2A) communication protocol.
• Design the Orchestration Layer: Lead the architectural design of our central semantic router hosted on AKS, that interprets user intent and orchestrates workflows across distributed, cross-platform agents.
• Enable Hybrid Innovation: Create "Golden Path" blueprints that leverage Databricks (Unity Catalog) for massive-scale data context while utilizing AKS to host complex, stateful agent swarms using frameworks like LangGraph or AutoGen.
• Drive Co-Creation & Governance: Partner with other platform teams (SAP, Salesforce) to co-architect complex cross-domain workflows (Tier 3-4) and serve as the primary technical authority for validating high-risk agent deployments.
• Ensure Zero-Trust Security: Embed security-by-design principles into our agent frameworks, ensuring every agent action is authenticated, authorized, and auditable within our Guardrail Engine.
  
  

Qualifications:

• Agentic Mastery: You possess deep conceptual and practical knowledge of agent design patterns (ReAct, Plan-and-Solve, Swarms, Hierarchical Teams) and know when to use each.
• Hybrid Infrastructure Expert: You are fluent in Azure AI Foundry for governance and model management, but you are equally comfortable architecting custom solutions on AKS to run open-source frameworks (LangChain/LangGraph) when managed services fall short.
• Data-Driven Architecture: Strong experience with Databricks, understanding how to expose complex data pipelines as "Tools" for agents and how to manage agent memory/state at scale.
• API & Protocol Design: Proven ability to design robust, event-driven architectures (CloudEvents, Kafka) and standard APIs (FastAPI, GraphQL, MCP) for distributed systems.
• Strategic Leadership: You can communicate complex architectural decisions to stakeholders and mentor senior engineers, fostering a culture of technical excellence.
• Tech Stack You Will Master:
• Orchestration: LangGraph, AutoGen, Semantic Kernel, Azure AI Foundry.
• Compute & Runtime: Azure Kubernetes Service (AKS), Docker, Azure Functions.
• Data & Memory: Databricks (Unity Catalog), Vector Databases (Qdrant/Pinecone), Redis.
• Languages: Python (Expert), TypeScript (Proficient).
• Standards: Model Context Protocol (MCP), OpenTelemetry.

 

Years of experience:

• 6