Opinions

Thoughts, research and analysis on Artificial Intelligence and how AI is reshaping the software development lifecycle.

6 September 2026 Observe, Protect, then Accelerate Two engineering leaders spent an hour on the agentic SDLC and barely mentioned code generation. Their early wins came from reading systems rather than writing them, and from fixing the queue before speeding up the thing that feeds it. Agents & Dev ToolingPolicy & Society 7 August 2026 LGTM, but did anyone actually read it? AI-authored code has roughly doubled the volume arriving for review while the humans reading it haven't moved, and unreviewed merges are climbing. Review isn't dying, it's losing its author, and rebuilding it takes more than a stricter LGTM. Agents & Dev Tooling 26 July 2026 How far would your agent go to complete its tasks? OpenAI's models broke into Hugging Face's infrastructure to retrieve the answers to the exam they were sitting. Optimise an agent hard enough and oversight becomes part of the attack surface. Safety & AlignmentAgents & Dev Tooling 20 July 2026 The Bottom Rung Is Missing AI is increasingly writing the routine, reviewable code that used to turn juniors into seniors. Headcount can rise while the apprenticeship quietly disappears. Productivity & LabourAgents & Dev Tooling 6 July 2026 The Off Switch You Don't Control When a US export-control directive suspended Claude Fable 5 for everyone, most enterprises had hedged the API. Almost none had hedged the business workflow built around it. Policy & SocietyInfrastructure & Compute 12 June 2026 No AI? We can't code any more! When one team's token cap hit, development ground to a halt entirely. If AI dependency is now a single point of failure, it deserves the same business-continuity thinking as any other critical vendor. Infrastructure & ComputeFunding & Markets 10 June 2026 Do developers know their AI-generated code is insecure? 70% of developers know AI-generated code is more vulnerable, and ship it anyway. The real question isn't whether they know, it's whether the gap is governance, overconfidence, or both. Safety & AlignmentPolicy & Society 16 March 2026 Are AI projects really failing? Failure rates from 45% to 95% aren't contradictory; they're different yardsticks measuring different things. The real question is whether the way we judge AI's success was ever fit for purpose. Productivity & LabourFunding & Markets 8 March 2026 Agentic AI Security is a new product category Claude Code Security and Codex Security don't scan code for known patterns, they reason about it. That shift from pattern-matching to hypothesis-forming changes what security tooling means for engineering leaders. Safety & AlignmentAgents & Dev Tooling 20 October 2025 Your developers are 55% faster, so why aren't you shipping any faster? AI code assistants make individuals faster, but Theory of Constraints says speeding up a non-bottleneck doesn't speed up the system. What your engineers do with the extra capacity matters more than the assistant itself. Productivity & LabourAgents & Dev Tooling