AI for Developers Where the LLM Helped, and Where It Didn’t: Designing a Hard Comparison UI While generating comparison concepts with a model, I had to change the team's approach; the AI couldn't have asked the question that mattered.
AI for Developers Permission Fatigue Is a Security Risk — Here’s How I Improved It in Claude Code Every permission prompt is supposed to be a checkpoint where a human looks at what the AI is about to do and decides whether it's safe.
Artificial Intelligence Bias and the Myth of the Objective Machine AI is genuinely good at like pattern recognition, speed, and repetition at scale. That also makes any bias it has dangerous.
AI for Developers Your Coding Agents Are Only as Good as Their Feedback Loop The size of the task you can hand to a coding agent is bound by the quality of its feedback loop and its ability to validate its work.
AI for Developers Tired of Reviewing Markdown? Make Yourself an HTML Spec Dashboard Drowning in markdown specs? Here's how I used AI to build a lightweight, regenerable HTML spec dashboard for faster, cleaner reviews.
AI for Developers A Hybrid Approach to Agentic Development with Local Models Frontier models are powerful, but increasingly expensive. Are local models a good option for helping to carry some of the workload?
Development Practices Human Habits for Leading Agentic Development Projects If you're leading an agentic development project and feel like it's taking some time to get your bearings, you aren't alone.
Delivery Practices The Knowledge That Dies When the Project Goes Live Without good documentation, the architect (me) becomes the primary source for context that should be institutional knowledge.
Artificial Intelligence Working with AI Is a Lot Like Building IKEA Furniture I have a soft spot for IKEA furniture because of how adaptable it can be. Working with AI has felt surprisingly similar.
Artificial Intelligence What “The Pitt” Gets Right About AI in Medicine If you want to see where clinical AI actually breaks, skip the vendor deck. Watch season two, episode six of "The Pitt."
AI for Developers Write Human Maintainable Code in the Age of AI Agents If code can only be modified safely by the original developer with the original agent setup, then the delivery is weaker than it looked.
AI for Developers AI This, Optimise That – But What Does Coding With It Actually Look Like? What does coding with AI look like in practice? From one developer to another, here’s my journey to find an agentic framework.
