--- title: "AI Agents Fundamentals" description: "Learn to design and build AI agents: agent architectures, communication protocols like MCP and A2A, multi-agent systems, and agent security." scfAsset: managed: false category: "guide" maintainers: - user: "can.celik1" external: true tags: ["STACKIT University", "Learning", "AI Agents", "MCP", "LangChain", "Multi-Agent"] source_url: "https://framework.stackit.cloud/data-and-ai/assetcontainer/stackit/ai-agent-fundamentals/" source_file: "docs/data-and-ai/assetcontainer/stackit/ai-agent-fundamentals.mdx" --- ## Course Overview This is a "zero to hero" course: it assumes no prior exposure to AI agents and builds from agent architecture (perception, reasoning, action) and reactive versus deliberative designs up through the frameworks used to actually build them — LangChain, AutoGen, CrewAI, and Semantic Kernel. Along the way it gives real weight to the emerging standardized protocols, Model Context Protocol (MCP) and Agent2Agent (A2A), that let agents call tools and talk to each other instead of being locked into one vendor's walled garden. The back half moves past single-agent demos into what production actually looks like: multi-agent orchestration and consensus, streaming responses over WebSockets and SSE for real-time interaction, and the alignment, security, and regulatory questions that come with letting an autonomous system take actions on your behalf. It closes with a capstone module where you design and document a complete agent solution end to end, rather than just reading about one — the same "own your build" philosophy carried through the rest of the AI Engineer path. ### What You'll Learn - Design agent architectures and choose between reactive and deliberative approaches - Build agents with frameworks like LangChain, AutoGen, or CrewAI, and connect them via MCP and A2A - Coordinate multi-agent systems with orchestration and consensus patterns - Implement streaming, real-time agent interactions over WebSockets and SSE - Evaluate the security, alignment, and regulatory risks of running agents in production ### Modules - AI Agent Foundations - Agent Communication Protocols - Agent Development Frameworks - Industry Applications and Case Studies - Multi-Agent Systems and Orchestration - Streaming and Real-Time Interactions - Ethics, Security, and Future Trends - Capstone Project and Professional Development