Letting AI refactor one legacy stylesheet
→
I pointed an AI model at a 1,200-line legacy Sass file and asked it to refactor everything to Tailwind. The useful part was mechanical syntax cleanup. The risky part was everything that depended on design, layout, and accessibility intent.
A tiny script that adds my internal links
→
A 40-line Python script adds internal links to my Ghost drafts via a shell alias. It is crude, fast, and only saves minutes, which is exactly why it works.
Three things I delete from almost every AI draft
→
When I use AI for first drafts, I spend most of my editing time deleting: fake origin stories, forced AI angles, and tidy morals the model invented for me.
The Hour I Lost to a Plausible AI Debugging Answer
→
A confident AI diagnosis of a React Native hydration issue sent me refactoring the wrong layer for 45 minutes. The real bug was a string "null" in the API.
What My Automated Archive Exposed About My Writing
→
Reading a batch of AI-written posts in my Ghost archive made the patterns obvious: recycled transitions, invented habits, generic intros, fake closure, and why even a few real notes change the result.
Where Cursor Stops and My Judgment Begins
→
Cursor is very good at filling in the boring middle. Recent client work showed exactly where that stops, and where experience still has to say no.
Whoop + AI: How I Predict My Recovery Windows Before I Burn Out
→
I wired my Whoop data into an AI pipeline that predicts my recovery windows before I feel tired. It changed how I schedule training, deep work, and rest.
Ethics Of AI In Frontend: Who Owns Your UI Components?
→
I have AI generate UI components almost every day now, but I do not trust any of them by default. This is how I think about responsibility and ethics when that code hits production.
Building A Personal Longevity Dashboard With bypass-longevity
→
How I wired the bypass-longevity AI lifespan predictor into my personal health dashboard, what I actually track, and how I keep it privacy-first and useful.
Biohacking With Whoop Data And AI: How I Actually Fixed My Circadian Rhythm
→
I stopped chasing Whoop recovery scores and started using AI to interrogate my sleep data. That is when my circadian rhythm finally started to make sense.
Productivity Hacking With AI: How I Let a Model Reorder My To‑Do List
→
I stopped manually grooming my to-do list and let a model reorder it using my own data. Here is how that actually works day to day, and what broke.
