<?xml version="1.0" encoding="utf-8"?>
<feed xmlns="http://www.w3.org/2005/Atom" xml:lang="en">
  <title>wito.dev blog</title>
  <subtitle>Notes on developer experience, NixOS, and agentic development.</subtitle>
  <link href="https://wito.dev/atom.xml" rel="self" type="application/atom+xml" />
  <link href="https://wito.dev/blog/" />
  <id>https://wito.dev/atom.xml</id>
  <updated>2026-11-15T00:00:00.000Z</updated>
  <author>
    <name>Witold Sosnowski</name>
    <uri>https://wito.dev/</uri>
  </author>
  <generator uri="https://astro.build/">Astro</generator>
  <entry>
    <title>Rust builds across worktrees, from 40 seconds to 2</title>
    <link href="https://wito.dev/blog/rust-build-time/" />
    <id>https://wito.dev/blog/rust-build-time/</id>
    <published>2026-11-15T00:00:00.000Z</published>
    <updated>2026-11-15T00:00:00.000Z</updated>
    <summary>I measured where a Rust build goes when every task gets its own worktree and its own devenv project. Copying target/ does nothing, Nix buys the cold start and loses the edit loop, and mbx does the opposite. The setup that works uses all three.</summary>
  </entry>
  <entry>
    <title>Working with multiple repositories, the easy way</title>
    <link href="https://wito.dev/blog/working-with-multiple-repositories/" />
    <id>https://wito.dev/blog/working-with-multiple-repositories/</id>
    <published>2026-11-15T00:00:00.000Z</published>
    <updated>2026-11-15T00:00:00.000Z</updated>
    <summary>The way we work changed dramatically. And with that, entirely new challenges popped up. In the recent months, I found myself working on projects sparnning multiple repositories. This was a real drag on productivity. So I decided to do something about it.</summary>
  </entry>
</feed>
