Design and configure effective AI agents
Connect agents to tools, information, and workflows
Build and maintain trusted AI-powered knowledge systems
Move beyond basic AI prompting and learn how to build AI systems that can take action on your behalf. Led by Harvard Professor Hanspeter Pfister, this short, project-based course gives you practical experience building, auditing, and deploying agentic AI workflows with Claude Cowork you can apply to your everyday work.
Across a series of tutorials, you’ll create a personalized morning briefing, automate note-taking, task-tracking, and email responses, and build a trusted local knowledge wiki. Along the way, you’ll learn the essential skills behind effective AI agents, from defining purpose and memory to connecting tools, managing permissions, evaluating sources, and maintaining reliable performance.
By the end of the course, you’ll have a practical toolkit for creating agentic workflows giving you the confidence to build new applications tailored to your own needs.
There are no formal prerequisites for this course; however, learners must have a Claude Pro subscription to fully participate in the course activities.
