Hey, I'm Loris.
Dad to Carla, the cocker in the photo, and a founder. Based in Italy, deep into tech and startups.
We're building a multi agent workspace that runs fully in the background for CUAs. Reach out if you're building something cool.
@kai_brokering Looks great and professional, love it. We just re-build our onboarding too for one app and it seems similar, happy to share it with you dm me
Everywhere on @X I see people posting about Jev from @typesafeai, and it looks great. Feels like I'm late even if it's been out two days. Starting to test it now, looks promising for computer use, judging by the demos going around.
@Pksingh900 Hey @Pksingh900, founder here too. We’re building a layer that let you run fleets of agents in background locally using your apps. Let’s connect!
@oli_so_no Hey @oli_so_no founder here too, would love to connect. We’re building a layer to run fleets of agents in background locally using your apps for you.
Don't take me wrong, it can work but it is not fully reliable. Apps can change their UI and the AX tree, this could broke your automations. I have experience even before Claude Code and gang with this matter and it tends to be fragile in real professional workflows.
Plus, you have a plently of apps that don't expose the full AX tree. Take the Adobe Suite (ex. Premiere Pro), it exposes the window and the menus, but the work surface is drawn by their own toolkit, so timeline, program monitor and most panels come through as one opaque element or unlabeled groups. The tree tells you a panel exists, not that there's a clip in it, where it starts or what's selected.
Honestly being able to really automate your apps in real professional workflows with computer use on your desktop needs still be fully solved
@the_rmoon Hey @the_rmoon, I'm a founder based in Italy. We're building a layer that runs fleets of agents on your desktop apps, on device and in the background, api or not. Let's connect!
@barnamarczali founder here too, from turin. 2 years automating legacy audio apps with no api taught us where desktop agents break. now we're building a layer that runs fleets of agents on your desktop apps, on device and in the background, api or not. we should connect @barnamarczali
@awlevin@typesafeai Interesting, will take a deeper look into your repo. Is this performing well overall as it seems? We'll start introduce this asap I guess
@georgevibing founder here too, from turin. 2 years automating legacy audio apps with no api taught us where desktop agents break. now we're building a layer that runs fleets of agents on your desktop apps, on device and in the background, api or not. we should connect @georgevibing
adobe suite (ex. premiere pro) is the clearest one. it exposes the window and the menus, but the work surface is drawn by their own toolkit, so timeline, program monitor and most panels come through as one opaque element or unlabeled groups.
the tree tells you a panel exists, not that there's a clip in it, where it starts or what's selected.
I’m working on this too, which is why your work caught my eye. would be good to chat sooner or later
We're working on a protocol that converts any ui into text at pixel level, this is the output scanning an app like DaVinci Resolve.
It lets us run agents on any app