<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Pydantic on Eugeniusz Zabłocki Blog</title>
    <link>https://ezablocki.com/tags/pydantic/</link>
    <description>Recent content in Pydantic on Eugeniusz Zabłocki Blog</description>
    <generator>Hugo -- gohugo.io</generator>
    <language>en-us</language>
    <copyright>© Eugeniusz Zabłocki</copyright>
    <lastBuildDate>Thu, 21 May 2026 10:00:00 +0200</lastBuildDate><atom:link href="https://ezablocki.com/tags/pydantic/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Conquering LLM Chaos: Guaranteed Structured Outputs with Pydantic and Ollama Grammar Sampling</title>
      <link>https://ezablocki.com/posts/sentinel-ai-part-2-structured-outputs/</link>
      <pubDate>Thu, 21 May 2026 10:00:00 +0200</pubDate>
      
      <guid>https://ezablocki.com/posts/sentinel-ai-part-2-structured-outputs/</guid>
      <description>In Part 1, we built a lightning-fast, concurrent highway for Sentinel AI—our local license compliance auditor. By leveraging LangGraph’s Map-Reduce pattern, parallel subgraphs, and database caching, we dropped the execution time for 60+ dependencies down to the speed of a single item.
But building a fast highway is only half the battle. When you populate that highway with reasoning models like deepseek-r1, you quickly realize they are incredibly talkative. They emit long, unpredictable streams of internal monologues, chain-of-thought steps, and conversational nuance.</description>
    </item>
    
  </channel>
</rss>
