Public preview
Iterative Refinement, Not Restart
Instead of rewriting the prompt, say "this part good. this part fix to X". 4 rounds = perfect output.
Goal
Leverage session memory, instead of restarting dialog each time.
Steps
-
1
Write first prompt, receive output.
Expected Outcome
Immediate
In 15 min you reach quality impossible via one big prompt rewrite.
Evidence base
School: Conversational AI / Human-in-the-Loop
Founders: Ethan Mollick · Anthropic (2024)
Mollick's "Co-Intelligence" 2024 describes this as "iterative co-piloting" — the key skill for working with LLMs.
Keywords
Frequently asked questions
How long does "Iterative Refinement, Not Restart" take?
This exercise takes about 15 minutes, practiced solo in a Creative format.
When will I notice the effect of "Iterative Refinement, Not Restart"?
In 15 min you reach quality impossible via one big prompt rewrite. In the short term: You learn dialog is an engineering process, not "single prompts".
Is "Iterative Refinement, Not Restart" evidence-based?
Yes — it draws on Conversational AI / Human-in-the-Loop (Ethan Mollick, Anthropic (2024)). Mollick's "Co-Intelligence" 2024 describes this as "iterative co-piloting" — the key skill for working with LLMs.
Is "Iterative Refinement, Not Restart" suitable for beginners?
Its difficulty level is: Beginner. No external tools required — it can be practiced directly inside the app.
Subscribe for full access
Subscribe to Plus for full access to this practice and the whole catalogue, or try first — free, no sign-in, no commitment.
7 complete practices (one per domain) are free forever, no sign-in.