<rss xmlns:atom="http://www.w3.org/2005/Atom" version="2.0">
    <channel>
        <title>AI &amp; Data - Category - ThanhLD Site</title>
        <link>https://ducthanh98.github.io/categories/ai-data/</link>
        <description>AI &amp; Data - Category - ThanhLD Site</description>
        <generator>Hugo -- gohugo.io</generator><language>en-us</language><lastBuildDate>Thu, 01 Feb 2024 10:00:00 &#43;0700</lastBuildDate><atom:link href="https://ducthanh98.github.io/categories/ai-data/" rel="self" type="application/rss+xml" /><item>
    <title>Generative AI: Chunking Strategies, Prompt Engineering &amp; AWS Bedrock</title>
    <link>https://ducthanh98.github.io/posts/genai/</link>
    <pubDate>Thu, 01 Feb 2024 10:00:00 &#43;0700</pubDate>
    <author>ThanhLD</author>
    <guid>https://ducthanh98.github.io/posts/genai/</guid>
    <description><![CDATA[Core concepts in Generative AI architectures, Retrieval-Augmented Generation (RAG) chunking strategies, Prompt Engineering techniques, and AWS Bedrock services.
RAG Chunking Strategies Chunking divides long documents into smaller segments for embedding and vector similarity search.
1. Fixed-Size Chunking (Standard) Concept: Chunks document into fixed token lengths (e.g. 300 - 512 tokens) with an overlap percentage (e.g. 10 - 20%). Pros: Fast, simple, predictable computation. Cons: Can cut sentences or paragraphs mid-thought, causing loss of semantic context.]]></description>
</item>
</channel>
</rss>
