I cannot possibly put into words how little I want to join your discord to get a single answer to a single technical question.

Here's a good rule of thumb: Too clever is dumb.
- Ogden Nash

More of this kind of thing, please.

Even among the generalized big-box LLMs, I still use them for different cases:
+ChatGPT is good for code
+Claude is good for debate/summary
+ Gemini is really good for tables, comparisons, then export to Docs

day in the life of a data engineer

It feels like a very enthusiastic dog on a leash is constantly tugging at your hand, asking you to go for a walk and won’t stop until you get your emotions into something you create

Looking for a *senior* dev to take the lead on outlines and other OSS initiatives at .txt. Experience with OSS libs is a huge plus. Send DM :)

it's crazy what kind of stuff people were able to paint during the renaissance without mspaint, much less photoshop and dall-e

“We’ve created a way to reduce hallucinations,” the whole LLM problem space is that they are vibe machines, that is literally their personality. If you want to use them, use them for tasks you don’t need six nines on or bound how often you’re willing to be wrong

New post on my search stack lately: Kagi, Marginalia, and, to some extent, Perplexity

vickiboykis.com/2024/04/25/how…

you wouldn’t last an hour in the asylum where they raised me

did you know you can just...download arxiv latex code? like they just put it there? mindbending lifehack for regenerating equations

I use pyenv almost daily and almost daily i forget how to set up and activate a new virtualenv with it akrabat.com/creating-virtu…

An absolutely fantastic way to increase this is to start a blog. Almost all the cool fun stuff in my professional life for me has come from doing stuff then blogging about it.

henrikkarlsson.xyz/p/search-query

I've just released llamafile v0.8 which features LLaMA3, Mixtral 8x22b, and Grok support. It goes 25x faster than ollama at running LLaMA3 70B on CPU. My new tensor multiplication kernels let llamafile eval MoE models 2x faster than llama.cpp github.com/Mozilla-Ocho/l…