Sam Carlton

Head of Engineering @TheACTHouse. Creator of @doesitarm and

[Sam Carlton @ThatGuySam]

@benln Happy to host one in Tulsa. We’ve got people super hungry for more of this

[Sam Carlton @ThatGuySam]

@alexalbert__ Hey, I’m curious if there’s anyway way to push along a Ambassador application for Tulsa.

I’ve been getting a flood of people that have been looking for a community to talk about Claude Code in and we’ve already got the infrastructure of Techlahoma.

[Gabe Nicholson @gabe_mohit]

Support Walter White

[Sam Carlton @ThatGuySam]

@Hesamation @lindavivah @robertnishihara @robertnishihara I’m sure there are reasons why this is almost impossible but I’m curious what the bottlenecks actually are.

Would it even be possible to get a good reply to a single message with downloading GBs of layers?

[Sam Carlton @ThatGuySam]

@Hesamation @lindavivah @robertnishihara @robertnishihara Great breakdown!

Working in Web Dev I’ve always suspected you might be able to eventually shard some aspect, like expert layers, across like sqlite-http vfs to be able to stream only the parts of the model you need but I just don’t know enough llms

[Sam Carlton @ThatGuySam]

Also you can try making it work yourself.

The code is completely open source and you can see all the details of how it works here: https://t.co/gFhFLxRd04

[Sam Carlton @ThatGuySam]

Have you ever gotten frustrated searching for the right emoji?

You know you've seen the emoji but you don't know the exact name.
Most search even on iOS using simple text matching:
🧵

[Sam Carlton @ThatGuySam]

This means our emoji search can match emojis to words it's never seen before. No exact match required.

It's not perfect, in fact, at the time of writing, it crashes a lot on mobile but it's a lot of progress for the constraints we're working under.