Thoughts on technology, AI, developer experience, and the occasional adventure.
In the past year, we've seen rapid advancement of model intelligence and convergence on agent scaffolding. But there's still a gap: agents often lack the domain expertise and specialized knowledge needed for real-world work.
AI articles next on my reading list
Why the data you see isn't always the whole story.
A profound reflection on interdependence and gratitude from Apple's co-founder.
simple protocol akin to HTTP without much opinion about payload
How does Claude think we react to hearing that?
tl;dr: "use cursor" to give up-to-date code documentation to your LLMs and AI code editors.
Knowledge cutoff is the AI model's "graduation day".
After weeks of confusion, I finally understand Model Context Protocols (MCPs). I break down what MCPs are, where they came from, and why they're a big deal for making AI models like Claude and Gemini more useful.
Confidence in AI = Less critical thinking. Confidence in self = More likely to question AI, verify info and think deeply.
A simple guide to HTTP status code categories and their perfect emoji representations.
Very large context windows, agents, and text action is gonna have an impact on the world at a scale that no one understands yet
What is a context window, and how much is One Million Tokens?
Kicking the tires on a new blogging system. Testing, testing, one, two... three.