Amir Khan is the father of the MAC Project — MuleSoft AI Chain — an ambitious open-source initiative designed to bring modern AI agent architectures into the hands of MuleSoft developers. The MAC framework makes it possible to build, test, and demo autonomous agents that reason, retrieve, and act — all within the MuleSoft ecosystem. With support for LLMs, vector databases, retrieval systems, and orchestration patterns, the project provides a powerful playground for exploring the future of integration intelligence.

In this session, Amir will share the why behind the MAC Project, how it was launched in response to internal disruption, and how it differs from initiatives like Agentforce. He’ll walk us through how MuleSoft developers can start prototyping AI-native workflows using tools like MCP (Model Context Protocol), A2A (Agent-to-Agent), and AGNTCY, and how to think about composing agents using concepts like embeddings, retrieval-augmented generation (RAG), and multi-agent coordination.

Whether you’re new to AI agents or already building with them, this talk will spark ideas and equip you with practical tools. Join us for a lively conversation with the architect behind MAC — a community-driven project that’s rethinking what agents can do in the world of integration.

Why attend?

If you’re a MuleSoft architect, engineer, or technically curious leader, this is a chance to hear directly from the creator of a fast-moving, future-focused open source project. You’ll walk away with a clear sense of how AI agents fit into modern integration patterns — and how to begin experimenting with them in your own projects. Expect practical insight, a few demos, and plenty of space for your questions.