- 582Better Experiments with LLM Evals — A funnel, not a forkSpotify Engineeringllm evalsa/b testingonline experimentationmetric calibration
- 584La Pedra Go Club in BarcelonaCassidoogobarcelonaboard gamestravel
- 585Don't Outsource the LearningAddy Osmaniai codingcognitive debtsoftware engineeringlearning mode
- 588Sort providers by cost, latency, or throughput on AI GatewayVercel Engineeringai gatewayprovider routinglatencycost
- 589Use native curl syntax with Vercel CLIVercel Engineeringvercel clicurldeployment protectionapi testing
- 590The Pulse: Did capacity shortages turn Anthropic hostile to devs?The Pragmatic Engineeranthropicclaude codegpu capacityxai