<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Large Language Models | Dr. Faried Abu Zaid - Personal Webpage</title><link>https://fariedabuzaid.github.io/tags/large-language-models/</link><atom:link href="https://fariedabuzaid.github.io/tags/large-language-models/index.xml" rel="self" type="application/rss+xml"/><description>Large Language Models</description><generator>Hugo Blox Builder (https://hugoblox.com)</generator><language>en-us</language><lastBuildDate>Mon, 25 May 2026 00:00:00 +0000</lastBuildDate><image><url>https://fariedabuzaid.github.io/media/icon_hu_982c5d63a71b2961.png</url><title>Large Language Models</title><link>https://fariedabuzaid.github.io/tags/large-language-models/</link></image><item><title>ibtidai: Adaptive AI Learning Studio</title><link>https://fariedabuzaid.github.io/project/ibtidai/</link><pubDate>Mon, 25 May 2026 00:00:00 +0000</pubDate><guid>https://fariedabuzaid.github.io/project/ibtidai/</guid><description>&lt;p>ibtidai is an AI-driven learning studio for self-directed, long-term skill building. The platform takes a broad learning goal, decomposes it into a structured sequence of focused lectures, and dynamically adapts the learning path based on the learner&amp;rsquo;s interests and observed progress.&lt;/p>
&lt;p>Key features include:&lt;/p>
&lt;ul>
&lt;li>&lt;strong>Dynamic curriculum planning&lt;/strong> — learning goals are automatically broken down into smaller, linked lecture units with a coherent progression&lt;/li>
&lt;li>&lt;strong>Adaptive learning paths&lt;/strong> — the next topic is selected based on demonstrated mastery, identified weak areas, and learner preferences&lt;/li>
&lt;li>&lt;strong>Spaced repetition&lt;/strong> — targeted question sessions and concept-level mastery tracking schedule review of weak areas over time&lt;/li>
&lt;li>&lt;strong>Knowledge graph&lt;/strong> — a visual concept map showing how learned topics relate and where gaps remain&lt;/li>
&lt;li>&lt;strong>Source-grounded lectures&lt;/strong> — content is generated with web grounding and source links are included for verification and deeper exploration&lt;/li>
&lt;li>&lt;strong>Bring-your-own-model&lt;/strong> — works with OpenAI, Anthropic (Claude), Google Gemini, and Groq; users supply their own API key for full cost control&lt;/li>
&lt;/ul>
&lt;p>The platform is particularly well-suited for technical fields that lack a clean off-the-shelf curriculum, such as game development, machine learning engineering, electronics, and applied mathematics.&lt;/p>
&lt;h2 id="early-access">Early Access&lt;/h2>
&lt;p>Available at &lt;strong>
&lt;/strong>&lt;/p></description></item></channel></rss>