<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:atom="http://www.w3.org/2005/Atom" version="2.0">
    <channel>
      <title>metravod.dev</title>
      <link>https://metravod.dev</link>
      <description>Personal blog about engineering, products and everything</description>
      <generator>Zola</generator>
      <language>en</language>
      <atom:link href="https://metravod.dev/rss.xml" rel="self" type="application/rss+xml"/>
      <lastBuildDate>Tue, 14 Apr 2026 00:00:00 +0000</lastBuildDate>
      <item>
          <title>What If We Build Agents Like a dbt Project</title>
          <pubDate>Tue, 14 Apr 2026 00:00:00 +0000</pubDate>
          <author>arttrek42@gmail.com (Artem Lyadov)</author>
          <link>https://metravod.dev/blog/what-if-we-build-agents-like-a-dbt-project/</link>
          <guid>https://metravod.dev/blog/what-if-we-build-agents-like-a-dbt-project/</guid>
          <description xml:base="https://metravod.dev/blog/what-if-we-build-agents-like-a-dbt-project/">&lt;p&gt;I came to agent development from data engineering, and once again, while putting together a typical structure on LangGraph, I started missing the declarative approach that many know well from dbt — where you describe what you &lt;em&gt;want&lt;&#x2F;em&gt; to do with the data, not &lt;em&gt;how&lt;&#x2F;em&gt;. And then a thought hit me: why not build my own agent framework that offers the exact same approach?&lt;&#x2F;p&gt;
&lt;p&gt;Plus, my least favorite thing about multi-agent systems on LangGraph is the mutable State. It quickly turns into an uncontrollable dumpster fire in RAM; you have to carefully update it every single time, and when something doesn’t work the way you want — you’re left printing things out and hunting for bugs. Maybe I’m just doing something wrong, but for me, these things always mean wasting extra mental energy.&lt;&#x2F;p&gt;
&lt;p&gt;Looking for a solution to this problem, I discovered the application of event-driven architecture in multi-agent systems, and later — event-sourced architecture. My main source of knowledge for this was a recent paper &lt;a rel=&quot;external&quot; href=&quot;https:&#x2F;&#x2F;www.alphaxiv.org&#x2F;abs&#x2F;2602.23193&quot;&gt;“ESAA: Event Sourcing for Autonomous Agents in LLM-Based Software Engineering” (Brito dos Santos Filho, 2026)&lt;&#x2F;a&gt;. Ultimately, I wrote my own framework — &lt;strong&gt;zymi&lt;&#x2F;strong&gt;.&lt;&#x2F;p&gt;
&lt;p&gt;But first things first.&lt;&#x2F;p&gt;</description>
      </item>
      <item>
          <title>Farewell to the Factory</title>
          <pubDate>Tue, 07 Apr 2026 00:00:00 +0000</pubDate>
          <author>arttrek42@gmail.com (Artem Lyadov)</author>
          <link>https://metravod.dev/blog/farewell-to-the-factory/</link>
          <guid>https://metravod.dev/blog/farewell-to-the-factory/</guid>
          <description xml:base="https://metravod.dev/blog/farewell-to-the-factory/">&lt;p&gt;In high school, I figured out what I wanted to be when I grew up pretty quickly—in the 9th grade, I ended up at a regional polytechnic olympiad held at the coolest tech university in Omsk. I only took 4th place, but my path was set right then and there.&lt;&#x2F;p&gt;
&lt;p&gt;Only my vision of this path was naive, as it turned out in practice. I thought I’d become some sort of Tony Stark who could build anything out of anything, as long as the desire was there. But in reality, it wasn’t so fun, and a design engineer in instrument-making spends way more time on boring paperwork that doesn’t really affect anything. Even when I eventually moved into IT, I kept running into the feeling that we are all just cogs in a machine.&lt;&#x2F;p&gt;
&lt;p&gt;And even when I started having an impact on the projects I was part of, I found myself at a crossroads: either you tone down your ambitions and keep doing hands-on engineering tasks, or you start discussing them while others actually do the work.&lt;&#x2F;p&gt;
&lt;p&gt;But now, with neural nets, everything has changed.&lt;&#x2F;p&gt;</description>
      </item>
    </channel>
</rss>
