Revision 176

Articles and updates:

• The Complete AI Experimentation Guide: Test, Compare, Validate & Ship Safely (link)

• The DIY platform trap that’s burning out engineering teams (link)

• The Problem with AI-Generated Post-Incident Reviews (link)

• You Shipped It Fast. But Did You Ship It Right? (link)

• On benchmarking (link)

• The Serverless Illusion: When “Pay for What You Use” Becomes Expensive (link)

• Humans aren’t fast enough for 4 9’s (link)

• The 28-Hour Meltdown: What Happened When AWS US-EAST-1 Overheated (link)

• From Proxy to Proxyless: Removing Envoy from Reddit’s Feed Serving Path (link)

• Semantics in Observability (link)

• Observing LLM Applications with OpenTelemetry (link)

• Kubernetes Autoscaling (link)