Current AI capabilities are great, but I don't feel we can use them to manage our inbox without compromising on our data. Also, in recent months, progress on open-source models has been impressive, with models getting smaller and more capable quickly.
That's why I had the idea of creating an email client from scratch — privacy-first and AI-native — that can run on a laptop without emails ever leaving your computer.
Over the last few months I've been working on EmailOps. I used Claude Code mainly to code the app, and I chose Rust and Tauri. It's my first app with these technologies, but I'm pretty impressed compared to others like Electron: the app is faster and more responsive, and the binary footprint is much smaller.
The app automatically classifies emails with tags using LLMs, drafts replies based on your past replies, and lets you chat with your inbox to search emails, summarize the content of specific emails, and so on.
I've open-sourced the project in case you want to have a look: https://github.com/emailops/emailops
Top comments (0)