Savvy
An assistant that whispers what to say during a meeting
What it does
Savvy is a macOS meeting assistant grounded in your own documents, not the open web. Point it at a folder and it builds a versioned brief per client. In the meeting it stays quiet, then speaks up for exactly three reasons: they asked a question your brief answers, someone crossed a red line you set, or you pressed Advice. Every card cites its source. Documents, indexes and transcripts stay on your Mac; only excerpts and the audio stream leave. Apple Silicon, macOS 13+. Open source, MIT.
Local-first macOS meeting assistant with source-grounded live guidance - jamalavedra/savvy
Savvy is a macOS meeting assistant. It prepares a brief from your own source documents, then offers source-grounded guidance during a live conversation. Preparation can be scoped to a specific client or kept general. Savvy keeps your documents and history on your Mac, but live transcription is not local. Read this before pointing it at anything confidential. Transcription requires a Deepgram or AssemblyAI account. Meeting audio is streamed to that provider as it is captured — there is no offline transcription mode in the current build. Savvy sets Deepgram's mip_opt_out=true so your audio is not used to train their models; check AssemblyAI's terms for their equivalent. Recommendations are…from github.com
Does the same job
all alternatives →
More dev tools this month
the category →
Open-source GTM skills for technical founders
Dev tools · 29d ago · gtmcofounder.com
Lettertrace▲375Track your AI visibility for free (using your own API keys!)
Dev tools · 25d ago · lettertrace.com
Launched alongside, August 2026
the whole month →- TL
Life & fun · 9d ago · louisabraham.github.io
- SA
Hello HN! I found that picking out plausible but diverse skin tones for my digital art and game development projects was kind of difficult, and I got curious about if there was a way to define a color space that made it easy. I've built a color picker and procedural generation algorithm based on the space as well as a bunch of other fun js features and demos throughout the page that use the equations. If you find it interesting, I have lots of explanations of how I built it and what properties the space has. The methodology might be a bit shaky, but hopefully the result is as helpful for…
Life & fun · Aug 2026 · toneyalexander.github.io
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 16d ago · simedw.com