Developer Advice
It takes time to learn how to work with AI
Better tools don't skip the learning curve. Even a year into working with AI, new adopters still have to struggle through the same messy process.
Finding balance with context switching
I tested my theory that those who context switch best produce the most value today. This is how I found my limit.
The era of build over buy
The buy vs. build needle has moved. Here's why I'm building directly on foundational tools instead of reaching for the next hyped-up abstraction.
The push toward async work
Early project stages need tight feedback loops while ideas take shape. Once there's real definition and confidence, productivity scales fast by shifting to async workflows.
AI can't read your thoughts
The prompt is the final step of a thinking process. AI tools can do a lot, but they can't replace the work of defining the unique and specific problems you are looking to solve.
Obsession: returning to the flow state of 2014
How modern AI coding tools haven't just made me faster, they have transported me back to the head-in-the-sand, deep-working feeling of my peak Ruby on Rails days.
You can't keep up with AI (and that's okay)
The pace of AI advancement is relentless. Learn how to stay productive without burning out trying to chase every new tool.
The era of the personal app
AI has advanced beyond silly little demos. It's now enabling anyone to build personal software catered to their unique needs and preferences.
When AI coding finally clicked
2025 has been a journey from struggling with the emergence of AI code generators to fully embracing it. From writing almost all the code to almost none of the code.
Abstract hard-coded values in your code
When JS frameworks don’t bring low-level structure recommendations on constants, configuration, and content, you’ll benefit from establishing your own conventions.