[brianswichkow]

Honestly, hard to say.

I gave it a ~2 min voice-to-text about my concept for the memo and, after retrieving relevant context via MCP—(1) my core context (claude.md, user.md, memory.md, etc. equvalents), (2) all contexts in my library from nearly a decade of writing about knowledge management, (3) knowledge of the Github repo for the project, and (4) Karpathy's post—it generated about 99% of the post (I changed a link and some formatting). That wouldn't have been possible without years of context, but this post was very much generated from such.

The beauty of the system is that knowledge, clarity, etc. compound.

[blahaj]

If you didn't write it, why should I or anyone read it?

This is AI slop. Opposed to your claim there is nothing clear about that page. It's really sad how ShowHN is spammed with slop that drowns out people's work. Please stop being part of that problem.

[brianswichkow]

While I can understand your distaste of AI slop and appreciate your feedback, if by your own admission you haven't read it, then it's hard to see your designation of my content as slop as justified. I welcome your feedback. If you choose to read it or review it with whatever AI agent you have created to help you synthesize, but I won't push you to do so if that's not something you feel called to do.

[blahaj]

I started reading it and found it to reek of AI slop. I couldn't figure out what I'm even reading about and found the text to be unstructured or mis-structured. It also seems to contain a lot of formatting mistakes making me believe that you didn't even care to proofread it.

Sorry to be so blunt, but I have no tolerance for AI a slop. ShowHN is for showing something you built with a lot of care, that people can try out. That can be an AI product. But it cannot be an AI slopped text post pitching your product.