Wan Leung Wong·Sep 28Knowing What’s Broken: Prometheus Metrics and an 88% Test Coverage SprintIn short: Once the pipeline started running unattended, logs were no longer enough. I added Prometheus metrics so failures were visible…
Ömer Faruk Yıldız·Aug 3Uncontrolled Speed is Fatal in AI(3)The tech world is full of successful launches, but lasting successes are exceptionally rare
InRaidubyShiva Ganesh Raidu·Jul 29ISO 42001 Clause 9.1 in Practice: Continuous Monitoring vs. the Quarterly Report
Massimo Brebbia·Jul 23AI Agents: Every Employee Is Becoming a ManagerLet me start with a statement that may sound deliberately provocative.
Alfred Vega·Jun 23The Hidden Cost of ProcrastinationOne of the things I’m currently learning in Month 2 of my internship is Advanced Meta Ads. Recently, our tasks included creating a company…
Wan Leung Wong·May 30When Your AI Agents Write Broken Code: A Four-Layer Accuracy SystemCode: github.com/wanleung/ai-dev-team
Wan Leung Wong·May 25One Watcher, Any Pipeline: Label-Based DispatchCode: github.com/wanleung/ai-dev-team
Wan Leung Wong·May 24Blocky Pipeline: Build Your Own Stage SequenceCode: github.com/wanleung/ai-dev-team
Wan Leung Wong·May 23Write the Test First: TDD Pipeline ModeIn short: The standard pipeline writes code first, then tests. This post covers a TDD mode that flips the order: QA writes tests before the…
Wan Leung Wong·May 16The LLM Relay: Pluggable Backends and Auto-FailoverIn short: The original pipeline was hardwired to GitHub Models. This post covers how I extracted every LLM backend into its own class with…
