7 min read
How the KV cache shapes AI agent design
Prompt caching only works when each request begins with the same tokens as the last. The design rules that follow for AI agents, and how to check them.
Most of what I write now is about building with AI agents and the context engineering around them. Older posts cover web performance, technical debt and leading engineering teams.
7 min read
Prompt caching only works when each request begins with the same tokens as the last. The design rules that follow for AI agents, and how to check them.
5 min read
An agent re-sends its whole conversation every turn. What that does to a bill, what a real project's numbers showed, and the five habits that changed it.
8 min read
AI coding tools are context engines. How the harness works, the practices that improve what it produces, and what to settle before a team adopts them.
6 min read
The mental model, the five levers that actually work, and the failure modes to watch for when you put an LLM into your daily workflow.
6 min read
RSA-OAEP encryption in the browser with the Web Crypto API, decrypted on a Java backend, with the key formats and padding that have to match.
7 min read
A high-level look at similarity search: embeddings, distance measures, the index structures that keep search fast, and what a vector database adds.
5 min read
Subscribe to RxJS observable updates from React using the useSyncExternalStore hook.
6 min read
Subscribe to Firestore live updates from React using the useSyncExternalStore hook.
1 min read
A worked example of applying the questions framework to a project you have just joined.
3 min read
Nine questions to ask when you join a new project, so you understand the product, the codebase and your own part in it before writing any code.
5 min read
What over-engineering costs a software product, why teams keep doing it, and how to keep a build aligned with what users actually asked for.
4 min read
Tech debt is a crucial concept for managers to understand, regardless of their technical background.
7 min read
This article will help you understand how to monitor and improve the overall performance of web applications.