Show HN: Autocrit – an agent loop that builds and tests web prototypes(github.com)
2 pointsby 4di11 hours ago1 comment
  • jsonfitzface9 hours ago
    The ambiguity doesn't go away with specs, it just gets deferred to when the code is written and you discover nobody wanted the thing you specified. If AI can shrink the loop between 'I think this is a problem' and 'someone who has this problem told me it isn't' that's genuinely useful, that's the part that kills most early products not the building itself.
    • 4di9 hours ago
      Well said.. yes, it’s more about validating hypotheses about what you’re trying to build.

      My own hypothesis is that we could create a rough representation of your end user and create an agent loop around that to try to validate. It won’t work for all stages of product development where you need humans to give you feedback but likely useful in the beginning stages