Agentic Integration & MCP Engineer

Capgemini is an AI-powered global business and technology transformation partner with nearly 60 years of heritage and 420,000 team members in over 50 countries. The Group reported 2024 global revenues of €22.1 billion. Capgemini commits to diversity, inclusion, fairness in employment, and responsible AI adoption.

• Over 4 years of experience in software engineering using Python, TypeScript, Java, or .NET including API design and integration with distributed systems.
• Hands-on experience with Agentic AI and Model Context Protocol (MCP), tool calling, agent orchestration, or similar agent integration approaches.
• Practical knowledge of AI frameworks such as Semantic Kernel, LangGraph, LangChain, Azure AI Foundry, OpenAI APIs, and Claude APIs.
• Strong understanding of REST, OpenAPI, OAuth 2.0, OpenID Connect, event-driven architecture, messaging, and secure system integration.
• Preferred experience with Retrieval-Augmented Generation (RAG), embeddings, vector databases, enterprise search, or semantic data access.
• Knowledge of LLMOps and observability tools like OpenTelemetry, Langfuse, and Arize Phoenix.
• Excellent proficiency in English.
• Commitment to building a Center of Excellence for Agentic Integration and secure engineering standards for AI integration.

• Design and develop MCP servers, agent tools, and secure connectors to enterprise systems.
• Integrate AI agents with APIs, databases, knowledge sources, messaging platforms, and SaaS products.
• Build agentic workflows using orchestration frameworks and event-driven integration patterns.
• Implement identity, authorization, secrets management, auditability, and guardrails for agent actions.
• Create evaluation, tracing, monitoring, and CI/CD practices for production-grade agent integrations.
• Contribute reusable components, reference architectures, and technical workshops for the AI and Integration communities.