Get live statistics and analysis of Douglas Correa's profile on X / Twitter

I'm returning to @elixirlang and testing Tidewave Web from @dashbit/@josevalim. It's surprising how it offers a unique coding agent experience, like a @v0 tailored to your stack, with point, click, and prompt features. It integrates with your backend, database, and more, enabling not only dev/ui tasks but also interacting with DB, etc. It brings new ideas for interacting with code via a coding agent. Excited to dive back into Elixir and explore it!

How to not love a language that allows you to schedule a task so easy and simple like this (ah, without any external dependency): #elixirlang #myelixirstatus

Well, I will repeat it again, @stitchbygoogle is awesome. I've generated almost every screens of a CRM application in less than one hour only prompting. And now I have all the images+html code along with the planning, that I'm doing with GPT-5 thinking + Opus to Claude code build it:

as I said, another day another cli. Install @aisdk cli and add this to your .zshrc. @GroqInc using gpt-oss-20b is so fast that you can just ask. cc @nishimiya

Besides I follow a lot of @elixirlang developers, if I looked to Twitter timeline, it seems there are only 2 programming languages: Python and Javacript

I made it! recordtab.io My first #buildinpublic endeavour. This is a simple "tab recorder" that allows you to record any chrome tab WITH sound, including your web meetings without any bot. Or you can record your entire screeen and talk over it, good for screencasts. I will continue to #buildinpublic but for now it is free!

I thought that @github copilot always use openAI models unless specifically asked, but I was wrong:

Did you know you can create a WebRTC app to share Chrome tabs with audio and send it right back to your server? I learned this while back and built a simple web meeting recorder with it. I discoverd this github.com/elixir-webrtc/… today. I think I will play with that project again #myelixirstatus

I just learned that @getpieces have a cli and I can easily ask: pieces ask --ltm "What I did yesterday?" and get a summary on everything I did yesterday since it is logging every activity Nice and useful via command line, since their mcp is not quite stable yet

Still using @zeddotdev and still liking it.

I just learned that the ACP protocol created by the @zeddotdev team has just opened the doors to deeply integrate Claude code into IDEs. Gemini CLI has already done it and seems promising. I think I will keep using Zed for a while.

I think @zeddotdev is promising, but its tab lags behind @cursor_ai. It seems to have the same issue as @jetbrains, trying to show autocomplete via IntelliSense alongside agent-based autocomplete. Maybe I'm missing a configuration, but this isn't a good developer experience.

as I said, another day another cli. Install @aisdk cli and add this to your .zshrc. @GroqInc using gpt-oss-20b is so fast that you can just ask. cc @nishimiya

I'm returning to @elixirlang and testing Tidewave Web from @dashbit/@josevalim. It's surprising how it offers a unique coding agent experience, like a @v0 tailored to your stack, with point, click, and prompt features. It integrates with your backend, database, and more, enabling not only dev/ui tasks but also interacting with DB, etc. It brings new ideas for interacting with code via a coding agent. Excited to dive back into Elixir and explore it!

Well, I will repeat it again, @stitchbygoogle is awesome. I've generated almost every screens of a CRM application in less than one hour only prompting. And now I have all the images+html code along with the planning, that I'm doing with GPT-5 thinking + Opus to Claude code build it:

Putting a spare @Hetzner_Online server to work testing @levelsio's raw-dogging strategy for quick idea validation. It's ideal for building and sharing niche prototypes with folks who'd actually care. Claude Code takes it up a notch—focused MVPs in minutes!

I think @zeddotdev is promising, but its tab lags behind @cursor_ai. It seems to have the same issue as @jetbrains, trying to show autocomplete via IntelliSense alongside agent-based autocomplete. Maybe I'm missing a configuration, but this isn't a good developer experience.

I thought that @github copilot always use openAI models unless specifically asked, but I was wrong:

I just learned that @getpieces have a cli and I can easily ask: pieces ask --ltm "What I did yesterday?" and get a summary on everything I did yesterday since it is logging every activity Nice and useful via command line, since their mcp is not quite stable yet

What happened with @elmlang ?

I should try @gleamlang it seems fun

I made it! recordtab.io My first #buildinpublic endeavour. This is a simple "tab recorder" that allows you to record any chrome tab WITH sound, including your web meetings without any bot. Or you can record your entire screeen and talk over it, good for screencasts. I will continue to #buildinpublic but for now it is free!