06 / INSIGHTS PRACTITIONER NOTES
Insights.
Practitioner notes on AI-native delivery, independent audits, and engineering risk.
01 What is an AI code review and why does AI-generated code need one?An AI code review is an independent audit of code written with AI assistance. Here is what it checks, how it differs from traditional peer review, and when to commission one. → 02 What are the problems with vibe coding in production?Vibe coding is genuinely fast and the speed is real. The problem is the gap between working and safe, and how invisible that gap is until someone finds it for you. The Lovable CVE and the Bolt key leaks show what that costs. → 03 Why is a free code audit from your dev shop not independent?A code audit is independent only when the reviewer has no commercial interest in the outcome. Free audits from the agency that built your platform, or the one that wants to rebuild it, fail that test before they begin. → 04 What is agentic development, and how is it different from AI coding tools?Agentic development is software delivery where AI agents do the manufacturing work under senior human oversight. The structural difference from AI coding tools is not the quality of the AI. It is where in the process the human shows up. → 05 What is AI-native software development and how does it differ from traditional engineering?AI-native software development designs the delivery process around what an agentic system can do reliably, with humans responsible for outcomes, architectural decisions, and validation. It is not a tooling choice. It is a process design choice. → 06 A Demo Isn't a FactoryEvery week someone sends me an AI coding demo that looks like magic. I'm impressed by the demos. I'm just rarely impressed by what happens in week three. → 07 What If the Best Engineer Never Writes a Line of Code?The engineering role has fundamentally changed. 80% is now judgement work. We're still hiring for the 20%. → 08 The 100x Question: Is the Iron Triangle Dead?What if speed and cost reduction weren't trade-offs? →