5 Comments
User's avatar
Modern Mom Playbook's avatar

Great article! The distinction between "AI can write the code, but it can't reason about systems" perfectly captures why so many AI-built projects feel brittle.

When you say "save working code every 30 minutes," are you using version control commits, or something else? I am curious how you balance granular checkpoints with keeping a clean commit history.

Expand full comment
Jenny Ouyang's avatar

Yeah, pretty much saving working code every 30 minutes.

When I’m using tools like Cursor, the model shows the changes it made, and I made it to require me to click accept to actually keep them. So part of “saving” is really about locking in those accepted changes before they vanish.

I also commit to Git pretty often, usually a few times within that 30 min window. AI makes editing fast, but deciding what to change takes longer. So frequent commits keep things safe without creating noise.

Sometimes I’ll batch a full feature into one commit for clarity, but most of the time I just commit whenever something works and feels stable.

Expand full comment
Modern Mom Playbook's avatar

Thanks, Jenny.

Expand full comment
Adria Kerr's avatar

Very interesting read! I have been starting to think that I want to learn more about AI and how to apply it to research. But I feel like I don't even know where to start

Expand full comment
Jing Xie's avatar

This is so good. Thanks for writing this!

Expand full comment