<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: Redjan Shabani</title>
    <description>The latest articles on DEV Community by Redjan Shabani (@redjan_shabani_019737b37e).</description>
    <link>https://dev.to/redjan_shabani_019737b37e</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F2163967%2Fbbb220d4-126a-4c47-a0e5-7f1408c3dc32.jpg</url>
      <title>DEV Community: Redjan Shabani</title>
      <link>https://dev.to/redjan_shabani_019737b37e</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/redjan_shabani_019737b37e"/>
    <language>en</language>
    <item>
      <title>Building a math-first poker engine in .NET</title>
      <dc:creator>Redjan Shabani</dc:creator>
      <pubDate>Sat, 31 Oct 2026 00:52:41 +0000</pubDate>
      <link>https://dev.to/redjan_shabani_019737b37e/building-a-math-first-poker-engine-in-net-2jo0</link>
      <guid>https://dev.to/redjan_shabani_019737b37e/building-a-math-first-poker-engine-in-net-2jo0</guid>
      <description>&lt;p&gt;This started as a solo engineering project, and it has grown into something much more interesting than a simple side experiment.&lt;/p&gt;

&lt;p&gt;I’ve been building a Texas Hold’em toolkit in C# focused on:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;bitwise data manipulation&lt;/li&gt;
&lt;li&gt;Monte Carlo equity estimation&lt;/li&gt;
&lt;li&gt;extensible game-state abstractions&lt;/li&gt;
&lt;li&gt;a low-level math-first design without relying on lookup tables&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;The core idea is not just to compute odds, but to build a reusable platform for poker analysis and strategy modeling with clean abstractions and strong performance characteristics.&lt;/p&gt;

&lt;p&gt;The project already includes:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;low-level card and rank primitives&lt;/li&gt;
&lt;li&gt;Texas Hold’em-specific logic&lt;/li&gt;
&lt;li&gt;CLI tooling for range-vs-range estimation&lt;/li&gt;
&lt;li&gt;benchmark projects for performance validation&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Current benchmark results for 10,000 Monte Carlo simulations:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;2 players: ~284.5 ms&lt;/li&gt;
&lt;li&gt;3 players: ~448.3 ms&lt;/li&gt;
&lt;li&gt;4 players: ~628.7 ms&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;This is still early, but the foundation is solid and there is a lot of space to evolve into deeper GTO-style work, more advanced equity modeling, and broader tooling.&lt;/p&gt;

&lt;p&gt;This project is now open for collaboration and contributors; if you are interested in:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;poker math&lt;/li&gt;
&lt;li&gt;algorithmic optimization&lt;/li&gt;
&lt;li&gt;C# / .NET&lt;/li&gt;
&lt;li&gt;simulation systems&lt;/li&gt;
&lt;li&gt;performance engineering&lt;/li&gt;
&lt;li&gt;open-source game-analysis tools&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;I’d love to connect and build this further together.&lt;/p&gt;

&lt;p&gt;Repo: &lt;a href="https://github.com/infokom-al/Infokom.Gaming.Poker" rel="noopener noreferrer"&gt;https://github.com/infokom-al/Infokom.Gaming.Poker&lt;/a&gt;&lt;/p&gt;

</description>
      <category>csharp</category>
      <category>dotnet</category>
      <category>texasholdem</category>
      <category>montecarlosimulation</category>
    </item>
  </channel>
</rss>
