Web Accessibility in the Age of AI Chat
How to structure blog content and embedded AI chat for maximum accessibility, and why good accessibility is now the foundation for AI-readable websites.
Recent Insights Page 2
AI and the Vanishing Cost of Intelligence
How artificial intelligence is relentlessly compressing the cost of intelligence and what it means when it effectively reaches zero.
Practical Code Reviews Over Fanatical Code Reviews
Why pragmatic, non-blocking code reviews lead to better software, happier teams, and faster delivery than fanatical perfectionism.
Performance per Watt in CPUs, GPUs & Data Centres: A 10‑Year Review
How computing efficiency has changed over the past decade across CPUs, GPUs, and at the data‑centre (infrastructure) level.
What Makes a Good Code Review?
How to provide constructive, clear, and respectful code review feedback that improves code and fosters team collaboration.
Just‑in‑time AI code generation: build the view you need now, not the whole app
How event‑driven, just‑in‑time AI code generation can reshape software delivery and reduce GPU pressure by materialising only the next required view at interaction time.
Average response time metrics for OpenAI GPT‑5 models (by reasoning level)
A practical look at typical end‑to‑end latencies when calling GPT‑5 models via the OpenAI API, how reasoning level affects time‑to‑first‑token and total completion time, and what you can do to measure and optimise it.
One Window, Many Packages: Use JetBrains Project Attach for Monorepos (works for separate repos too)
Monorepo-first guide to Project Attach in JetBrains IDEs: attach apps/packages from a workspace (pnpm/yarn/npm, Nx, Turborepo) for focused tooling, while still supporting separate repositories.