Rafa

😅 I spent a few weeks building a macOS app and learned more about development than in years. Hapi v1.0.1 just shipped 🚀 Voice Notes 100% Local (FREE), plus Meeting recordings with diarization (Soon available). 🔗 Link in first comment 👇

@NocodeTalks @mattdotroberts Here: chat.whatsapp.com/ByLqHbfVy167h1…

@mattdotroberts How to join community?

i run a 180+ ai curious whatsapp community. Claude takes the chats, rips out a digest for whatsapp and online Creates a live dashboard to understand behaviour Makes sure the community get the sauce not all the noise explained it here and see the comments if you want to hit the digest

@mattdotroberts Yeiiii!!!

@rafa_maker vibe community building ain’t it

@mhrescak I am 😅

@rafa_maker Are you doing actual diarization locally? Sweet.

Granola was overkill for my markdown note setup, so I had Claude build a fully local transcription CLI with a skill that works great and composes well with others. Read more: hrescak.com/notes/transcri…

@mattdotroberts And Q2 plan 🚶‍♂️

Company all hands is go

@picsoung @hackbarna @itnig Won’t miss the next one!

This was yesterday A room full of hackers discovering the power of skills for @hackbarna “skill a thon” at @itnig in Barcelona 🤩 People built skills to help meal planning, help with cold emails, or run a11y tests.

@mattdotroberts Tomorrow I have to show you things 😬

I’m a bookmark boy. Who’s with me?

@mattdotroberts @cjpedregal We actually have this set up already for speakhapi.com - connected to Claude seamlessly

@cjpedregal Ok back to the drawing board @rafa_maker It does show there’s a need

There are some tweets out there saying that Granola is trying to lock down access to your data. Tldr; we are actually trying to become more open, not closed. We’re launching a public API next week to complement our MCP. Read on for context. A couple months ago, we noticed that some folks had reversed engineered our local cache so they could access their meeting data. Our cache was not built for this (it can change at any point), so we launched our MCP to serve this need. The MCP gives full access to your notes and transcripts (all time for paid users, time restricted for free users). MCP usage has exploded since launch, so we felt good about it. A week ago, we updated how we store data in our cache and broke the workarounds. This is on us. Stupidly, we thought we had solved these use cases well enough with our MCP. We’ve now learned that while MCPs are great for connecting to tools like Claude or chatGPT, they don’t meet your needs for agents running locally or for data export / pipeline work. So we’re going to fix this for you ASAP. First, we’ll launch a public API next week to make it easier for you to pull your data. Second, we’ll figure out how to make Granola work better for agents running locally. Whether that’s expanding our MCP, launching a CLI, a local API, etc. The industry is moving quickly here, so we’d appreciate your suggestions. We want Granola data to be accessible and useful wherever you need it. Stay tuned.

@mattdotroberts Hahaha yes sir!

Nice take Coordinating humans in a group is a pain in the ass. Coordinating 2 founders to the next shiny toy. Easy Ain’t that right @rafa_maker

This is going to be wild if you are in Barcelona Thursday next week 👇

@elenayuscula 100%

por esto cuando me hablan ahora de "crear contenido" (por VISIBILIDAD reasons) realmente solo quiero llorar. son super dark times, yo solo quiero hablar de las cosas que me gustan y que eso aiude a mi negocio no puedo con el performativismo continuo, we have to go back 😢

Try it here 👉 speakhapi.com

Tested Hapi at the first @AnthropicAI event in Barcelona — seated far from the stage, barely hearing the speakers. ✅ Hapi caught every word anyway. Filtered filler, fixed errors, made sense of it all in real time. Very nice feeling when things work! 👇

Download it here 👉 speakhapi.com

🎉 Big Hapi update for Hapi: → Qwen 3.5 models now power content repurposing (huge quality jump) → Redesigned automation UI → Self-correction system for AI outputs → Faster, lighter performance Still 100% local. Still one-time purchase. Free while Beta #BuildInPublic

@elenayuscula Jejeje en realidad tienes razón 🙌

@rafa_maker 😱 la documentación *nunca* sobra, Rafa 👀

mi mayor miedo cuando delego procesos tan grandes como el del rediseño de aprendenotion.com es si luego seré independiente en el mantenimiento, así q cuando le he pedido a Claude una guía con todo y me ha escrito hasta los prompts para pedirle cosas, he LLORADO

Hi all! anyone has successfully integrated new Qwen3.5 models in any OS app? I’m trying to do it for speakhapi.com however they are multimodal and the mlx is not yet compatible. But I’ve seen others do it… any ideas? thanks!

@adrgrondin Hi Adrien! Just downloaded localAI and will give a try now. May I ask how did you use the queen new models? I’m trying to upgrade to the for my app, and i can’t make it work since they are multimodal and I only need text. Any suggestion is welcome! Thanks!!

I’m working on many improvements to Locally AI, mainly performance first to make the app faster and more reliable. Improvements to the Qwen 3.5 models are also coming, as well as a Mac update to add the models. Thanks to everyone who downloaded the app.

@Valen Eres un máquina!

Desde hoy VA360 es la única comunidad en la que sus CoProjects tienen una suscripción de Claude Code Max incluida para todos los participantes sin coste extra. 👇🏼 🔥