Andrés Felipe Quintero Moreano

I wish there were laptops with built-in split keybooards (ignore my terrible photoshopping skill issues)

@shuttle_dev No need for a Box

In case you were wondering... Zed is proudly built the hard way—from scratch, starting with an empty repo. Our team *created* Electron, but we no longer believe that the ultimate editor can be created by forking a web browser. (Or by forking a fork of a web browser 🙃)

@dillon_mulroy Weekend work? Or weekend open source? Two very different types of people

this is what happens when you keep your engineers out of pointless meetings and give them trust and autonomy. they ship.

@shahzadahmad246 I like fish

what's your go-to terminal/command-line tool for daily tasks?

>linux kernel developer >no syntax highlighting >no LSP >one terminal window, no tmux >literally just rawdogging terminal and barebones vi >writes USB driver from scratch in just 3h what's your excuse anon?

@MaverickAlbert5 3838

Developers, apart from localhost:3000, what other port do you use?

@JosiahParry did you benchmark the new impl on your computer?

@not_shreyaaaa Is this rage bait?

@kennedymwavu @JosiahParry I'm looking more at 2x. But we'll see

@AndyQuinteroM @JosiahParry yeah, need to cut off that 1 extra ms :)

@kennedymwavu @JosiahParry I still think there is a lot of room left to improve.

@JosiahParry @AndyQuinteroM i ask again, when do you stop? when do you acknowledge that it's fast enough?

@the_codewala Mys-Quel

How do you pronounce your Mysql? -M-Y-S-Q-L -MY-S-Q-L -MY-Sequel

@JosiahParry Did we write the fastest implementation there is for encoding geohash in R?

@JosiahParry check your PRs

@JosiahParry That would be my way of doing it. I will use unsafe 😤

@JosiahParry Give me the dataset, I’ll get you a faster implementation.

Hyperthreading is Dead. Major CPU manufacturers (including Intel) are getting ready to abandon SMT hyperthreading completely. It's not as crazy as you think. In fact, Apple, Qualcomm, and Samsung have *never* used Hyperthreading. But why? 🧵

Are you tired of hard or inefficient routers for Shiny? Check out this video where I teach you how to deploy multiple Shiny apps on a single container with faucet youtu.be/hQEdbrb2iTc

@Hasen_Judi I think it’s due to the fact that most ML work is not done by software engineers. Doing ML in a statically typed compiled language does not seem like the best of trade offs for ML experimentation.

The fact that most open source ML work is done in python is kinda sad for me. There's nothing about the field that requires python or even makes python preferable. It makes local setup and experimentation needlessly painful.