Page 6 of 6

Reflection 3

How This Reflection Works

Your Facilitator will pull from these questions to lead a cohort-wide discussion. If your AI assistant is building during a Challenge and you have a free moment, these also make great team conversation starters while you wait.

What You Built

• What's the most ambitious feature your team added in this challenge? What made you choose that direction?
• What's the difference between an application that shows generic information and one that responds to today's actual conditions? Why does context-aware output feel different to a user, and what does it take to make that shift?
• Is there something in your app right now that you'd genuinely find useful if someone were using it for real?

What You Practiced

• Lesson 3 used the "house-sitter note" metaphor for project context: you write down what someone would need to know to take care of your project without calling you. What belongs in that note? How would starting every AI conversation with that context already in place change the way you work, compared to re-explaining your project each time?
• A project context file isn't static; the project evolves and the context should evolve with it. What kinds of changes to a project should trigger an update to the context file? What happens if the context file gets stale?
• Lesson 3 described the "table of contents, not the whole book" principle for context files. What's the risk of putting too much in a context file? What's the risk of putting too little? How do you decide what earns a place in the file and what stays out?
• Think of the context file as a team agreement about how AI should work on your project. If your team had to agree on a single document that shapes every AI interaction, what would make it in and what would get cut? How does framing it as a "team agreement" change what you'd include?

How You Worked

• Think about the progression across the three phases: first learning to prompt in chat, then moving to a real environment with real data, then building with persistent context. How would you expect a team's working style to change across those stages? What capabilities does each stage unlock?
• Was there a moment in this challenge where one teammate's idea took the project in a direction nobody else had considered?

Looking Ahead

• You've been building something that connects to real data and serves a real purpose, but it only lives in a development workspace. What's the gap between "works on my machine" and "someone else can use this"? What would need to happen to get something like this in front of real users?