Info This post is auto-generated from RSS feed Hacker News. Source: OTelBench: AI struggles with simple SRE tasks (Opus 4.5 scores only 29%)