I’ve been working on a project built with Next.js. Like most projects that want to be discoverable by humans, it needed a blog.

I didn’t want to:

run a headless CMS (Strapi, Contentful, etc.)

build and maintain a full blog UI in Next.js

deal with SEO tradeoffs from putting content on a separate subdomain

So I went with a slightly unfashionable but practical solution: WordPress.

The goal was simple:

Main app at yoursite.ai (Next.js)

Blog at yoursite.ai/blog/...

WordPress doing what it’s good at: writing, previews, publishing

No SEO penalty from split domains

There are a lot of questions online about this setup, and surprisingly little concrete guidance. This is the approach that worked for me.

Thanks for sharing. We do the same thing at BiggerPockets but via Cloudflare workers to do the proxying.

Hi, this is a short Saturday morning project.

I usually want to learn how big repos work, but looking at the latest commit is not worth it. They more often than not contain a lot of yadda yadda and edge cases that aren't helpful when learning.

So I like to go to the very first commit and step forward.

This tool helps me do that, hope you find it useful as well!

What's the difference with lowering the temperature?

I think ideally you want the whole path to be the most probable path, which is not likely to be the same as taking the most probable token at each step.

It's not remotely practical to select the most probable path but you can do a little bit of search a few tokens at a time.

Openai owns the intelligence until it doesnt, and the open source model is good enough

To all the commenters asking how is this product different from another product:stop asking that question! Its a month old! instead why don't you chime in with your idea to help this founder out on how he can differentiate?

The peers asking them how this differentiates from the many, many other solutions for this are doing him an enormous favor. If he knows, it may prompt him to improve the messaging on his site. If he doesn't, creating a crisp answer to that question will improve his chances of success.

An example of the output would be nice

Super interesting, please do show what the other trees (collatz) that you mention look like

Coming for v2

OP here, Is there a way to figure that out?

(Not OP) I can think of a convoluted and expensive pair-wise comparison method, but I hope there's also a way to figure this out during the application of principal component analysis in a way I don't understand.

Edit: I'm thinking it can't be done without experimentation on the embedding model.

Edit2: Ah, even that might not yield results, because as the basis is derived interstitially through computation, there's no guarantee the features of the final coordinate system will have any accessible relationship to those of the initial basis.