Hacker News: hangrymoon01https://news.ycombinator.com/user?id=hangrymoon01Hacker News RSShttps://hnrss.org/hnrss v2.1.1Fri, 17 Apr 2026 09:40:24 +0000<![CDATA[New comment by hangrymoon01 in "Launch HN: Relvy (YC F24) – On-call runbooks, automated"]]>Re: savings - it depends on the use case. For example, one of our users set up a small runbook to run a group-by-IP query for high-throughput alerts, since that was their most common first response to those alerts. That alone cuts out a couple of minutes of exploration per incident and removes the variability of the agent deciding what data to investigate and how to slice it.

In our experience, runbooks provide a consistent, fast, and reliable way of investigating incidents (or ruling out common causes). In their absence, the AI does its usual open-ended exploration.

]]>Thu, 09 Apr 2026 18:53:51 +0000https://news.ycombinator.com/item?id=47708050hangrymoon01https://news.ycombinator.com/item?id=47708050https://news.ycombinator.com/item?id=47708050<![CDATA[New comment by hangrymoon01 in "Launch HN: Relvy (YC F24) – On-call runbooks, automated"]]>Re: custom harnesses, imo maintaining them can be time consuming especially when things are changing very fast with AI. Bringing up a prototype is easy but a robust harness that handles the edge cases needs time and effort.

]]>Thu, 09 Apr 2026 18:39:07 +0000https://news.ycombinator.com/item?id=47707793hangrymoon01https://news.ycombinator.com/item?id=47707793https://news.ycombinator.com/item?id=47707793<![CDATA[New comment by hangrymoon01 in "LLM Costs of AI investigating production alerts"]]>Interesting. Nice to see someone publishing actual cost numbers.

]]>Mon, 16 Mar 2026 14:04:34 +0000https://news.ycombinator.com/item?id=47399212hangrymoon01https://news.ycombinator.com/item?id=47399212https://news.ycombinator.com/item?id=47399212