I'm building ata, an open source research and coding agent. Started as an internal tool, open sourced it a few weeks ago. It helps me combine and understand concepts across multiple papers, pull the best ideas from each, and turn them into code. It uses Treesitter and Lsp for multi-repo code understanding.
The focus is on the interfaces: there's a reading view that keeps you locked in without drowning in information, and a voice mode with karaoke-style word highlighting. Also built an iOS app for research on the go, releasing that soon.
npm install -g @a2a-ai/ata https://github.com/Agents2AgentsAI/ata/
> Look at your app, describe a sequence of changes out loud, and watch them happen in front of you.
The problem a lot of times is that either you don't know what you want, or you can't communicate it (and usually you can't communicate it properly because you don't know exactly what you want). I think this is going to be the bottleneck very soon (for some people, it is already the bottleneck). I am curious what are your thoughts about this? Where do you see that going, and how do you think we can prepare for that and address that. Or do you not see that to be an issue?