<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>GenAI | Tuhin Sharma</title><link>https://tuhinsharma.netlify.app/tags/genai/</link><atom:link href="https://tuhinsharma.netlify.app/tags/genai/index.xml" rel="self" type="application/rss+xml"/><description>GenAI</description><generator>Hugo Blox Builder (https://hugoblox.com)</generator><language>en-us</language><lastBuildDate>Mon, 10 Mar 2025 14:15:00 +0000</lastBuildDate><image><url>https://tuhinsharma.netlify.app/media/icon_hu55b84836e614877e119cbfa37f6d5a66_1386708_512x512_fill_lanczos_center_3.png</url><title>GenAI</title><link>https://tuhinsharma.netlify.app/tags/genai/</link></image><item><title>RAPTOR - A Smarter Way to Retrieve and Use Information in AI</title><link>https://tuhinsharma.netlify.app/blogs/raptor/</link><pubDate>Mon, 10 Mar 2025 14:15:00 +0000</pubDate><guid>https://tuhinsharma.netlify.app/blogs/raptor/</guid><description>&lt;h2>Description&lt;/h2>
&lt;p>Large language models (LLMs) like ChatGPT and GPT-4 are incredibly powerful, but they struggle to keep up with new information and understand long, complex documents. Traditional retrieval methods pull short chunks of text, often missing important context. This is where RAPTOR (Recursive Abstractive Processing for Tree-Organized Retrieval) comes in — a new way to retrieve and summarize information efficiently. In this article, we’ll break down what RAPTOR is, why it’s better than traditional methods, and how to implement it with a hands-on example.&lt;/p>
&lt;p>&lt;a href="https://medium.com/@tuhinsharma121/raptor-a-smarter-way-to-retrieve-and-use-information-in-ai-fd3cb68a6f2f">&lt;p>&lt;strong>Read more..&lt;/strong>&lt;/p>&lt;/a>&lt;/p></description></item><item><title>Building a Biomedical Question-Answering System Using RAG</title><link>https://tuhinsharma.netlify.app/blogs/rag-openai/</link><pubDate>Mon, 03 Mar 2025 14:15:00 +0000</pubDate><guid>https://tuhinsharma.netlify.app/blogs/rag-openai/</guid><description>&lt;h2>Description&lt;/h2>
&lt;p>With the explosion of biomedical research and literature, extracting relevant information efficiently has become a challenging task. Large Language Models (LLMs) like GPT-4o-mini, combined with a Retrieval-Augmented Generation (RAG) pipeline, offer a promising approach for building robust question-answering (QA) systems.&lt;/p>
&lt;p>&lt;a href="https://medium.com/@tuhinsharma121/building-a-biomedical-question-answering-system-using-rag-and-openai-llm-b9c3502fd287">&lt;p>&lt;strong>Read more..&lt;/strong>&lt;/p>&lt;/a>&lt;/p></description></item></channel></rss>