marviel

Thanks for sharing this blog post is a great speculative deep-dive.

it mentions native multimodality somewhere in either the Arxiv or post -- seems like it might handle it well?

nice! Can't agree more on Vibe Speccing.

I wrote more about Spec Driven AI development here: https://lukebechtel.com/blog/vibe-speccing

how do you account for the success of reasoning models?

I agree these things don't think like we do, and that they have weird gaps, but to claim they can't reason at all doesn't feel grounded.

have often wanted something like "liquid types", but didn't know what it was called! Thanks for this

there is now I suppose! ;)

Yeah it isn't a panacea but it has afforded me less frustration than the alternative of jumping straight in.

Since what I've been working on is just a mash-up of other people's ideas

Totally, I find most work I do, if I'm honest, is in this bucket. LLMs are pretty good at "filling in the gaps" between two ideas like this

I'm almost to this one in my read through! I'm excited to get to the "information age" arc

I totally agree!

this is the idea behind my recent post actually[1] where I recommend people use AI to write specs before they code. If all you have to do is a human is edit the spec, not write it from scratch, you're more likely to actually make one.

[1] https://lukebechtel.com/blog/vibe-speccing

def cleanup inactive specs over time.

it's only useful if it'll be used!

Yeah just learned about this!

The md files are actually pretty great for shareability, versioning, and picking up where you left off.

I need to try Claude Code's plan mode! Definitely feels useful.

Totally. Yeah I think your approach is a solid take!

Thanks for the writeup!

I talked about a similar, but slightly simpler workflow in my post on "Vibe Specs".

https://lukebechtel.com/blog/vibe-speccing

I use these rules in all my codebases now. They essentially cause the AI to do two things differently:

(1) ask me questions first (2) Create a `spec.md` doc, before writing any code.

Seems not too dissimilar from yours, but I limit it to a single LLM

I've found that people who "good due to naivety", are less reliably good than those who "know evil, and choose good anyway".

i need this in ableton

yeah -- I saw it's built on "open source foundations", do you know what project this is?

Thanks for sharing!

Which embedding types did you try? I'm surprised that embeddings weren't able to take you further with this.

There are two good books that explore this concept from different angles:

"The Emotion Machine", by Marvin Minksy (the AI view)

&

"The Art of Empathy", by Karla McLaren (the Internal / Emotional View)

"Wired For Story" is a great book that discusses an alternative set of "Story Atoms" which can be combined to create unique, compelling stories.

what do you dislike about it?

mermaid probably.

We've found a lot of success with this approach at Reasonote -- would love to hear about how others solve the same issues.

every day the culture grows nearer

Thank you!

Thanks!

Thanks for the feedback! I'll put these in as GitHub issues and try to work them out once I'm back from the gym :)

what are your personal evals?

o3 still drives really well for me.

source?

"Want to get a job? [COLLEGE] is having a 0% discount -- only $200,000 a year! Don't you want to have a place to live? Go horribly in debt, pick a degree that may not matter, all at an age where your Brain has not yet developed fully!"

etc

Not perfect now, but adequate in some domains. Will only get better.