<?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: Karthikeyan</title>
    <description>The latest articles on DEV Community by Karthikeyan (@karthikeyan_bcfdbd86e07c0).</description>
    <link>https://dev.to/karthikeyan_bcfdbd86e07c0</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%2F4201524%2F7295554c-8b1d-4ca8-9789-d88640679fe3.png</url>
      <title>DEV Community: Karthikeyan</title>
      <link>https://dev.to/karthikeyan_bcfdbd86e07c0</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/karthikeyan_bcfdbd86e07c0"/>
    <language>en</language>
    <item>
      <title>GhostLog 👻 – AI-Powered Error Investigator Using Gemma 3 1B</title>
      <dc:creator>Karthikeyan</dc:creator>
      <pubDate>Fri, 30 Oct 2026 18:23:25 +0000</pubDate>
      <link>https://dev.to/karthikeyan_bcfdbd86e07c0/ghostlog-ai-powered-error-investigator-using-gemma-3-1b-1325</link>
      <guid>https://dev.to/karthikeyan_bcfdbd86e07c0/ghostlog-ai-powered-error-investigator-using-gemma-3-1b-1325</guid>
      <description>&lt;h2&gt;
  
  
  What I Built
&lt;/h2&gt;

&lt;p&gt;I built GhostLog, an AI-powered error log investigator for developers.&lt;/p&gt;

&lt;p&gt;Users can paste Python or JavaScript errors and receive an AI-generated investigation containing the error summary, root cause, explanation, step-by-step solution, corrected code example and prevention tips.&lt;/p&gt;

&lt;h2&gt;
  
  
  Code
&lt;/h2&gt;

&lt;p&gt;GitHub Repository:&lt;/p&gt;

&lt;p&gt;&lt;a href="https://github.com/karthikeyan24-kk/GhostLog" rel="noopener noreferrer"&gt;https://github.com/karthikeyan24-kk/GhostLog&lt;/a&gt;&lt;/p&gt;

&lt;h2&gt;
  
  
  How I Built It
&lt;/h2&gt;

&lt;p&gt;I used Python, Flask, HTML, CSS, JavaScript, Ollama and Gemma 3 1B.&lt;/p&gt;

&lt;p&gt;The Flask backend receives the error logs and sends them to the locally running Gemma model through Ollama's API.&lt;/p&gt;

&lt;p&gt;The model analyzes the error and returns a structured explanation.&lt;/p&gt;

&lt;p&gt;I also implemented input validation, connection error handling, timeout handling and automated tests.&lt;/p&gt;

&lt;h2&gt;
  
  
  Why Does Open Innovation Matter?
&lt;/h2&gt;

&lt;p&gt;I chose Gemma 3 1B because it is an open-weight model that can run locally using Ollama.&lt;/p&gt;

&lt;p&gt;This means users can analyze their programming errors without sending their logs to an external AI API.&lt;/p&gt;

&lt;p&gt;Once the model is downloaded, local inference can work without an internet connection.&lt;/p&gt;

&lt;p&gt;Open innovation gives developers the freedom to experiment with models, understand how their AI tools work and build applications without depending on paid cloud AI services.&lt;/p&gt;

&lt;h2&gt;
  
  
  The Spirit of It
&lt;/h2&gt;

&lt;p&gt;Every developer encounters mysterious errors that seem to appear from nowhere.&lt;/p&gt;

&lt;p&gt;GhostLog investigates those hidden problems and helps developers understand the ghosts inside their code.&lt;/p&gt;

&lt;p&gt;My goal is to make debugging easier, especially for beginners who are still learning programming.&lt;/p&gt;

&lt;h2&gt;
  
  
  Prize Categories
&lt;/h2&gt;

&lt;p&gt;Best Use of Gemma&lt;/p&gt;

</description>
      <category>devchallenge</category>
      <category>hf26challenge</category>
    </item>
  </channel>
</rss>
