Collaboration on AI workflows requires more than just sharing boards; it demands a culture of constructive critique. When you review someone else's process, your goal is to help them refine their logic and improve their results. This requires moving beyond simple reactions and offering insights that are specific, actionable, and grounded in clear reasoning.
In this lesson, we examine how to leave comments that actually help. We break down the difference between critiquing the processâhow a workflow is structuredâand critiquing the outputâwhat the AI actually generated. By mastering these distinctions, you become a more effective collaborator and help your team build more robust, reliable AI workflows.
Assignment
Review a peer's shared board. Leave at least three comments: one addressing the process, one addressing the AI output, and one suggesting a specific, actionable alternative to a node. Ensure all your feedback is grounded in clear reasoning.
Learning Objectives
- Understand the difference between feedback on the process and feedback on the AI output.
- Learn to provide specific, actionable, and grounded feedback on shared boards.
- Identify and avoid common feedback pitfalls like vague praise and scope creep.
Specific, Actionable, Grounded
Useful feedback points to a specific node, suggests a clear alternative, and explains the reasoning behind the suggestion.
Process vs. Output Feedback
Distinguish between commenting on how a workflow is mapped (the process) and what the model actually produced (the output). Both are valuable but require different approaches.