Ask HN: What's the costliest mistake you've made with LLM agents in production?
3•jairooh•1h ago
Comments
jairooh•1h ago
Ours was a LangChain agent we deployed to automate customer research. No budget cap, no monitoring, no approval gates. It ran a recursive loop overnight querying external APIs — we woke up to a $600 bill and zero useful output.
That incident is literally why we built AgentShield (useagentshield.com). Runtime risk scoring on every agent action, cost tracking per run with kill switches, and human approval gates before high-risk decisions.
Curious what others have run into — cost overruns seem common but prompt injection incidents in production are the ones that scare me most.
mtmail•1h ago
If more and more Ask HN just promote new services I loose motivation to answer.
appreciatorBus•1h ago
4 disguised promotion submissions in 2 weeks including this one, 0 contributions on any other topic except your own product. Give it a rest, go buy some ads or something.
jairooh•1h ago
That incident is literally why we built AgentShield (useagentshield.com). Runtime risk scoring on every agent action, cost tracking per run with kill switches, and human approval gates before high-risk decisions.
Curious what others have run into — cost overruns seem common but prompt injection incidents in production are the ones that scare me most.