3 posts tagged with "python"

AI agents are getting more capable, but they are also getting increasingly expensive in a way that is easy to overlook.

A single agent run may repeatedly pay for a system prompt, conversation history, retrieved context, memory, tool schemas, tool outputs, intermediate reasoning, and the final response. When the workflow becomes multi-step, token consumption compounds quickly.

Most benchmarks ask a binary question: did the agent complete the task?

I wanted to ask a second one:

How many tokens did it actually need to succeed?

That question led me to build Suffice, an open-source toolkit for measuring the minimum token budget an AI agent needs to complete a task while preserving a required level of quality.

A few weeks ago I published a post about building ncmds — a zero-configuration, Markdown-driven documentation site builder I was running on Flask + Python. The post got some traction, a handful of stars on GitHub, and a few contributors poking around the code. That was great.

Then I quietly rewrote the entire backend in TypeScript.

This post is the honest story of why.

A few months ago I found myself in a familiar situation: I needed to document a side project quickly. I had used Docusaurus before, loved it in theory, but every time I bootstrapped a new doc site I ended up fighting Node.js toolchains, React component boilerplate, and config files before I had written a single line of actual documentation. MkDocs was simpler, but it felt dated and lacked the AI-native features I increasingly wanted in my workflow.

So I did what every engineer does when the existing tools annoy them enough — I built my own. The result is ncmds (No Code Markdown Documentation Sites), and this post walks through the thinking, the architecture, and the lessons learned.