⌘K
Change language Switch ThemeSign In
Narrow Mode
Diagrams as a Key to Effective Engineering Plan Review in AI Development
Diagrams as a Key to Effective Engineering Plan Review in AI Development
 ### Garry Tan@garrytan
/plan-ceo-review and /plan-eng-review skills basically gets you 90% of the way to done most of the time
the secret in plan-eng-review is that you can always ask for a diagram, and the act of creating the diagram (user flow, data flow, branching, edge cases) shakes out the bugs
#### dex
@dexhorthy · 17h ago
damn this is so good and encapsulates everything I've been seeing/saying in the last few months
- a spec that is sufficiently detailed to generate code with a reliable degree of quality is roughly the same length and detail as the code itself
- so don't review those things, just review the code at that point, if you care enough about that level of abstraction
- unless you're vibing side projects or prototypes (yes, even zero-to-one software), you ABSOLUTELY SHOULD care about the code at that level of abstraction
- you need to find SOME way to get more leverage over coding agents though, because just reading all that code is a pain, esp when a lot of it is slop
- the default/dare-i-say-decel way is to go back to "i own the execution, and give little things to the agent, check it along the way"
- the accel-but-safe-way is to find something - NOT A SPEC (the word "spec" is broken anyway) - NOT 3 INVOCATIONS OF AskUserQuestion - that lets you resteer the model before it slops out N-thousand LOC Show More
14
238
44.8K
Mar 18, 2026, 3:58 AM View on X
19 Replies
7 Retweets
107 Likes
14.7K Views G Garry Tan @garrytan
One Sentence Summary
Garry Tan highlights that asking for diagrams during engineering plan reviews is a crucial 'secret' to identify bugs early, especially relevant in the context of AI-generated code challenges.
Summary
Building on a quoted tweet that critiques overly detailed specs for AI-generated code, Garry Tan, CEO of Y Combinator, offers a practical solution: leveraging '/plan-eng-review' skills. He emphasizes that the 'secret' to success in this process is to always ask for diagrams (user flow, data flow, branching, edge cases). The act of creating these diagrams helps to 'shake out the bugs' before extensive code is written, effectively addressing the challenge of managing AI-generated code quality and steering models before they produce large amounts of 'slop'. This provides a concrete, high-leverage method for ensuring quality in AI-assisted software development.
AI Score
80
Influence Score 37
Published At Today
Language
English
Tags
Engineering Review
AI Development
Software Quality
Diagrams
Garry Tan HomeArticlesPodcastsVideosTweets