Module 5: Building your custom agent – Part 3: Uncovering unconscious bias

Module 5 overview

In this workshop, you’ll learn to build AI agents designed specifically to support the content development workflow. We’ll use structured reference data and carefully designed prompting to handle more complex, task-specific outputs. We’ll explore practical examples, including agents that can adapt to specific authorial voices, assist with generating assessment items aligned to curriculum standards, and repurpose existing content for different formats.

Learning outcomes

By the end of this module, you will be able to:

• Upload and structure reference data to train an agent on a specific voice or curriculum

• Understand and implement ‘chain-of-thought’ prompting within an agent in order to handle complex content-related tasks

• Build an agent that functions as a tool for auxiliary content creation (e.g., to produce quizzes, summaries, glossaries)

• Evaluate the output of custom agents against standards