Going to AWS re:Invent 2026? Join us at booth 1871 for demos, giveaways, and more.
Meet with us

Signal vs. Spend: Building Cost-Aware Observability at Slack

Slack engineers Emma and Steven traced a single noisy log line to a 311-billion-log-a-day pipeline with no volume limits or cost attribution. In this O11yCon 2026 talk, they show how a custom logging MCP server and Claude-powered PR reviews cut Slack's logging costs 30%+ company-wide for under $500 in AI spend.