How I keep up with AI on X without living there
I built X Analyst over the weekend to help me keep up with interesting AI stuff on X without spending so much time separating signal from noise.
Writing
Thoughts on AI, engineering, and building products.
How to give agents context that does not fall apart
Why filesystem-first context architectures work, where they break down, and how the same idea can be implemented with structured object spaces using small, composable primitives instead of bespoke tools.
Documentation is entering the runtime
As AI systems begin to consume documentation at runtime, documentation shifts from external reference to operational input. This article explains why in this setting runtime-coupled documentation becomes unavoidable, even more so for on-prem and regulated deployments, and why this is an architectural boundary, not a tooling choice.
Beyond vectors: Why semantic search breaks on engineering documents, and how to fix it
In domains where every letter, number, and logic gate matters, "close enough" in semantic similarity can be dangerously wrong. This article shows why vector search often fails on engineering content and describes a hybrid, structured approach that handles technical detail far more reliably.
No, AI does not make us stupid
Using AI for software engineering and writing doesn't simplify my job - instead I move quicker, handle more decisions, and deal with more moving parts. As a result, work becomes more intense, and I cover more ground and learn more than ever before.
The matrix of knowledge, or, why Excel is way cooler than you may think
Matrices—simple grids of rows and columns—have become one of humanity's most powerful tools for organizing complex information, from tracking terrorists to creating music, managing investments, and analyzing DNA. In this article I explore how versatile the matrix approach to organizing information is, and why it turns out to be such a great fit for AI.
Hallucinations, interpretations, and unknowables
While hallucinations are often discussed in the context of LLMs, there are other phenomena that are often confused with hallucinations.
Should you fine-tune your LLM?
'Fine-tuning an LLM is when you take a general LLM and train it further on a domain-specific
Why counting results doesn't make sense for modern search
'Modern search engines use smart matching that finds similar meanings,
AI-first data storage
AI-first data storage means storing data in a way so they can be used by an AI system. AI systems here particularly are LLMs (large language models).
What is RAG? What are tool calls? And why is this important?
RAG, or retrieval-augmented generation, is a method for combining LLMs, or large language models, with outside knowledge. This outside knowledge could be news, websites, R&D publications or sensor data, for example. And tool calling enables LLMs to use external tools.