Follow the development of Aethas, a proactive AI executive assistant. Technical deep-dives, architecture decisions, and honest progress updates from a solo founder.
When I started building Aethas, I started with a purely local-first approach. Everything would run on the user's machine: their data, their embeddings, their conversations. No cloud required. This was great for learning Tauri and iterating quickly. I...
Aethas indexes local files like Obsidian vaults, or markdown folders for context-aware conversations. This happens at startup, when users manually trigger a re-index, and continuously via file watchers. After some use I found it annoying to guess if ...
Aethas indexes Obsidian vaults and other sources, making that content searchable via RAG. Users can @mention files from their indexed sources. But what about that PDF someone just sent you? Or a code file from a different project? The friction: copy ...
Back in 1996, when I was just six years old, I watched Star Trek for the first time. I can remember sitting on the floor of my father's study in front of his cathode-ray television. We watched all of the Star Trek TV shows and films that had come out...