News for hackers. Stuff that matters.

A comment on Launch HN: Prized (YC S26) – Let non-engineer staff build secure internal tools. Parent comment

Re:Launch HN: Prized (YC S26) – Let non-engineer staff build... by hudsongr on Thursday July 30, 2026 @03:44PM (#49114802)
We assume LLMs can always be fooled, so the judge is just one step of the verification process. The deterministic layers, per-tool db roles, read-only-by-default integrations, query timeouts, are the ones an admin can configure. The judge's rules aren't exposed to admins today. I like your idea of letting admins add their own rules or knowledge to the judge, we should add that.
[ Parent | on HN ]
Re:Launch HN: Prized (YC S26) – Let non-engineer staff build... by Arshad-Talpur on Thursday July 30, 2026 @04:18PM (#49115208)
One of the core issue here is LLMs are probabilistic not deterministic, even with set of rules or context , llms are inherently built to make best guess, so my question is how you are ensuring that LLMs are following rules?
[ Parent | on HN ]
Re:Launch HN: Prized (YC S26) – Let non-engineer staff build... by marinoseliades on Thursday July 30, 2026 @06:32PM (#49116711)
You're right, and we don't. The judge is best-effort screening not enforcement. Enforcement is deterministic with per-tool Postgres roles, proxy-injected creds, host allow-list, human approval on destructive writes. Those hold whether the judge is right or wrong.
[ Parent | on HN ]