{"id":4458,"date":"2026-05-05T14:33:55","date_gmt":"2026-05-05T17:33:55","guid":{"rendered":"https:\/\/beontech.wpengine.com\/?p=4458"},"modified":"2026-05-05T14:45:42","modified_gmt":"2026-05-05T17:45:42","slug":"ai-powered-querying","status":"publish","type":"post","link":"https:\/\/beon.tech\/blog\/ai-powered-querying\/","title":{"rendered":"AI-Powered Document Search with LangChain, ChatD &amp; NotebookLM"},"content":{"rendered":"\n<p>AI is changing how we interact with information. One of the most practical shifts is how we can now <strong>ask questions directly to our documents<\/strong>: PDFs, books, manuals, and other unstructured sources. Without manually searching page by page.<\/p>\n\n\n\n<p>In this webinar, <strong>Matias Bustamante<\/strong> walks through a hands-on tour of <strong>three different approaches<\/strong> to querying PDFs with AI:<\/p>\n\n\n\n<ol class=\"wp-block-list\">\n<li><strong>LangChain + OpenAI (Python)<\/strong> using a Retrieval-Augmented Generation (RAG) workflow<\/li>\n\n\n\n<li><strong>ChatD + local LLMs<\/strong> (via Ollama) to query PDFs offline<\/li>\n\n\n\n<li><strong>NotebookLM<\/strong> (Google) powered by <strong>Gemini<\/strong>, with a no-code interface and strong citations<br><\/li>\n<\/ol>\n\n\n\n<p><strong>&#x1f3a5; Watch the Full Webinar<\/strong><\/p>\n\n\n\n<iframe loading=\"lazy\" width=\"560\" height=\"315\" src=\"https:\/\/www.youtube.com\/embed\/N6vhmbcwIlw?si=EVxzRUWAPs75e-2X\" title=\"YouTube video player\" frameborder=\"0\" allow=\"accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share\" referrerpolicy=\"strict-origin-when-cross-origin\" allowfullscreen><\/iframe>\n\n\n\n<p><em>Prefer to watch? The complete session is above.<\/em><em><br><\/em><em> Prefer to read and replicate the implementations? Keep scrolling.<\/em><\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>What You\u2019ll Learn<\/strong><\/h2>\n\n\n\n<p>By the end of this guide, you will:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Understand what an <strong>LLM<\/strong> is (in simple terms) and why it works<\/li>\n\n\n\n<li>Learn what <strong>RAG<\/strong> is and when it improves accuracy<\/li>\n\n\n\n<li>See how <strong>indexing \u2192 retrieval \u2192 generation<\/strong> works in document Q&amp;A<\/li>\n\n\n\n<li>Build a PDF Q&amp;A flow using <strong>LangChain + OpenAI<\/strong><\/li>\n\n\n\n<li>Compare results against <strong>local models<\/strong> (Mistral \/ DeepSeek) using ChatD<\/li>\n\n\n\n<li>Use <strong>NotebookLM<\/strong> to query multiple documents with citations, no code required<\/li>\n\n\n\n<li>Understand real-world tradeoffs: accuracy, cost, privacy, and usability<br><\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>Quick Tour of Recent AI Updates<\/strong><\/h2>\n\n\n\n<p>Before going hands-on, Matias briefly highlights a few AI stories and what they signal:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>DeepSeek<\/strong>: positioned as a major launch with claims about efficiency and training improvements<\/li>\n\n\n\n<li><strong>Baidu (Ernie)<\/strong>: new versions announced, with a focus on cost and capability improvements<\/li>\n\n\n\n<li><strong>Arizona Supreme Court<\/strong>: experimented with <strong>AI avatars<\/strong> to communicate announcements publicly<\/li>\n\n\n\n<li><strong>Italian newspaper<\/strong>: published an edition created with AI support, with journalists curating the output<br><\/li>\n<\/ul>\n\n\n\n<p>These examples reinforce the theme of the webinar:<strong> AI is moving fast and practical workflows matter.<\/strong><\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>Two Key Concepts: LLMs and RAG<\/strong><\/h2>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>What Is an LLM?<\/strong><\/h2>\n\n\n\n<p>An LLM (Large Language Model) can be thought of as a system that <strong>predicts the next word<\/strong> in a response based on patterns learned from large-scale training data.<\/p>\n\n\n\n<p>The breakthrough that made modern systems fast and effective is transformer-based attention, enabling stronger context handling and faster generation.<\/p>\n\n\n\n<p>Common commercial uses include:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Customer support<\/li>\n\n\n\n<li>Content drafting<\/li>\n\n\n\n<li>Software development assistance (e.g., copilots)<br><\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>What Is RAG?<\/strong><\/h2>\n\n\n\n<p>RAG (Retrieval-Augmented Generation) is a technique to <strong>improve accuracy<\/strong> by giving the model <strong>relevant external information<\/strong> at answer time.<\/p>\n\n\n\n<p>This matters when:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Your model\u2019s training data may be outdated<\/li>\n\n\n\n<li>Your knowledge lives in private documents (manuals, internal PDFs, books, policies)<br><\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>The 3 Steps of RAG<\/strong><\/h2>\n\n\n\n<ol class=\"wp-block-list\">\n<li><strong>Indexing<\/strong>: split documents into chunks and store them for retrieval<\/li>\n\n\n\n<li><strong>Retrieval<\/strong>: search for the most relevant chunks (often via vector similarity)<\/li>\n\n\n\n<li><strong>Generation<\/strong>: pass the retrieved chunks to the LLM so it answers grounded in the source<br><\/li>\n<\/ol>\n\n\n\n<p>Result: <strong>more accurate, document-grounded answers<\/strong>.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>Approach 1: Query PDFs with LangChain + OpenAI (Python)<\/strong><\/h2>\n\n\n\n<p>This first approach is a classic RAG implementation using Python.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>Requirements<\/strong><\/h3>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Python 3.9<\/li>\n\n\n\n<li>pip<\/li>\n\n\n\n<li>OpenAI API access (paid usage via credits)<br><\/li>\n<\/ul>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>What LangChain Does<\/strong><\/h3>\n\n\n\n<p>LangChain is an open-source framework that simplifies building LLM applications, especially for workflows like retrieval, chaining, prompt templates, and evaluation.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>The Workflow Used in the Demo<\/strong><\/h3>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Load the PDF (via a PDF loader)<\/li>\n\n\n\n<li>Split text into chunks (with overlap to avoid losing context)<\/li>\n\n\n\n<li>Create embeddings for chunks<\/li>\n\n\n\n<li>Store vectors in a vector database (Chroma in the demo)<\/li>\n\n\n\n<li>Retrieve relevant chunks for a question<\/li>\n\n\n\n<li>Generate an answer using an LLM (GPT-4 used in the session)<br><\/li>\n<\/ul>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>Demo Questions<\/strong><\/h3>\n\n\n\n<p>Matias tests the pipeline with two documents and questions:<\/p>\n\n\n\n<ol class=\"wp-block-list\">\n<li><strong>Clean Code (Robert C. Martin)<\/strong><strong><br><\/strong> Question: <em>Are comments good or bad in code?<\/em><em><br><\/em> The response reflects the book\u2019s stance (e.g., \u201ccomments as a necessary evil\u201d and emphasis on expressive code\/refactoring).<br><\/li>\n\n\n\n<li><strong>A car manual (Chery Tiggo 7 Pro)<\/strong><strong><br><\/strong> Question: <em>What\u2019s the purpose of the auto-hold system?<\/em><em><br><\/em> The answer includes both explanation and operational steps.<br><\/li>\n<\/ol>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>Key Note About Privacy \/ Data Sent<\/strong><\/h3>\n\n\n\n<p>Only <strong>relevant chunks<\/strong> are sent to the LLM for answering (not the entire book\/PDF).<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>Pros \/ Cons (as presented)<\/strong><\/h3>\n\n\n\n<p><strong>Pros<\/strong><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Flexible: you can swap models (OpenAI, Anthropic, Gemini, etc.)<\/li>\n\n\n\n<li>Highly customizable: prompts, multi-question scripts, evaluation<\/li>\n\n\n\n<li>Strong baseline for building real products<br><\/li>\n<\/ul>\n\n\n\n<p><strong>Cons<\/strong><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Requires setup + code<\/li>\n\n\n\n<li>Paid API usage<\/li>\n\n\n\n<li>Output formatting and multi-doc UX require extra work<br><\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>Approach 2: Query PDFs Locally with ChatD + Ollama<\/strong><\/h2>\n\n\n\n<p>This approach uses <strong>ChatD<\/strong> (desktop app) plus <strong>Ollama<\/strong> to run LLMs locally (e.g., Mistral, DeepSeek).<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>Requirements<\/strong><\/h3>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Ollama running locally<\/li>\n\n\n\n<li>ChatD installed<\/li>\n\n\n\n<li>A local model downloaded<br><\/li>\n<\/ul>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>Why This Approach Matters<\/strong><\/h3>\n\n\n\n<p>Local setups can be valuable when:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>You want to avoid sending data to external APIs<\/li>\n\n\n\n<li>You need offline access<\/li>\n\n\n\n<li>You\u2019re testing multiple models quickly<\/li>\n<\/ul>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>Demo Comparison<\/strong><\/h3>\n\n\n\n<p>Matias repeats the same questions against different local models.<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>With <strong>Mistral<\/strong>, the \u201cClean Code\u201d answer is described as more generic\/misleading compared to the RAG-grounded response.<\/li>\n\n\n\n<li>With <strong>DeepSeek<\/strong>, the answer is more detailed, but still may not quote\/ground as well as the first approach.<br><\/li>\n<\/ul>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>Limitations Noted<\/strong><\/h3>\n\n\n\n<ul class=\"wp-block-list\">\n<li>You\u2019re restricted to local models only<\/li>\n\n\n\n<li>Typically works with <strong>one document at a time<\/strong><\/li>\n\n\n\n<li>If you need to query a \u201clibrary\u201d of documents together, it can feel limiting<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>Approach 3: Query PDFs with NotebookLM (Gemini, No-Code)<\/strong><\/h2>\n\n\n\n<p>This is the simplest workflow: upload documents into <strong>NotebookLM<\/strong> and ask questions through a web UI.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>What Makes NotebookLM Stand Out<\/strong><\/h3>\n\n\n\n<ul class=\"wp-block-list\">\n<li>No local install<\/li>\n\n\n\n<li>No coding<\/li>\n\n\n\n<li>Works well with multi-document \u201csources\u201d<\/li>\n\n\n\n<li>Shows <strong>which chunks were used<\/strong> to answer (strong for trust and verification)<\/li>\n\n\n\n<li>Can generate an <strong>audio\/podcast-like<\/strong> overview (English-only noted)<\/li>\n<\/ul>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>Demo Outcome<\/strong><\/h3>\n\n\n\n<p>For both test questions (\u201cClean Code\u201d and the car manual), NotebookLM produced answers that Matias describes as <strong>more in-depth and better grounded<\/strong> than the other options, especially because it surfaces citations\/chunk references.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>Tradeoffs Mentioned<\/strong><\/h3>\n\n\n\n<ul class=\"wp-block-list\">\n<li>NotebookLM is a Google experiment and could change or be discontinued<\/li>\n\n\n\n<li>If usage spikes, response time or availability could vary<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>When Will RAG Still Matter With Huge Context Windows?<\/strong><\/h2>\n\n\n\n<p>A question from the audience: if context windows grow massively, will RAG become obsolete?<\/p>\n\n\n\n<p>Matias\u2019 take: RAG will still matter in cases like:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Sensitive company documents<\/li>\n\n\n\n<li>Internal knowledge that shouldn\u2019t be uploaded<\/li>\n\n\n\n<li>Local\/private setups where retrieval pipelines remain useful<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>Key Takeaways<\/strong><\/h2>\n\n\n\n<p>This webinar shows three practical ways to \u201cchat with your PDFs,\u201d each suited to different needs:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>LangChain + OpenAI (Python)<\/strong>: best for building robust, productizable pipelines<\/li>\n\n\n\n<li><strong>ChatD + local models<\/strong>: best for privacy\/offline experimentation (with workflow limits)<\/li>\n\n\n\n<li><strong>NotebookLM (Gemini)<\/strong>: best for no-code, multi-doc querying with strong grounding<br><\/li>\n<\/ul>\n\n\n\n<p>The core concept underneath all of them is the same: <strong>RAG-style retrieval improves reliability when answers must be tied to source documents<\/strong>.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>F<\/strong>AQs<\/h2>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>With bigger context windows, will RAG become outdated?<\/strong><\/h3>\n\n\n\n<p>No. RAG remains valuable for sensitive\/internal documents and controlled retrieval workflows.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>What would you choose for day-to-day use: OpenAI, local, or NotebookLM?<\/strong><\/h3>\n\n\n\n<p>Matias\u2019 preference based on the demo: <strong>NotebookLM<\/strong>, due to depth and grounding\/citations.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>How should a data scientist start learning LLMs if they\u2019re new to the field?<\/strong><\/h3>\n\n\n\n<p>Start with <strong>prompt engineering fundamentals<\/strong> and learn how prompt quality changes output quality.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>Do OpenAI servers receive the entire document?<\/strong><\/h3>\n\n\n\n<p>In a RAG workflow, the model typically receives <strong>only relevant chunks<\/strong>, not the full document.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>AI is changing how we interact with information. One of the most practical shifts is how we can now ask questions directly to our documents: PDFs, books, manuals, and other unstructured sources. Without manually searching page by page. In this webinar, Matias Bustamante walks through a hands-on tour of three different approaches to querying PDFs<a class=\"read_more_linkk\" href=\"https:\/\/beon.tech\/blog\/ai-powered-querying\/\">&#8230;<\/a><\/p>\n","protected":false},"author":52,"featured_media":4461,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"_sitemap_exclude":false,"_sitemap_priority":"","_sitemap_frequency":"","_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"footnotes":""},"categories":[491],"tags":[],"class_list":["post-4458","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-webinars"],"acf":[],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v27.8 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>AI-Powered Querying for PDFs | Webinar by Matias Bustamante<\/title>\n<meta name=\"description\" content=\"Learn three practical ways to use AI-powered querying for PDFs: LangChain + OpenAI (Python), ChatD with local LLMs, and Google NotebookLM.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/beon.tech\/blog\/ai-powered-querying\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"AI-Powered Querying for PDFs | Webinar by Matias Bustamante\" \/>\n<meta property=\"og:description\" content=\"Learn three practical ways to use AI-powered querying for PDFs: LangChain + OpenAI (Python), ChatD with local LLMs, and Google NotebookLM.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/beon.tech\/blog\/ai-powered-querying\/\" \/>\n<meta property=\"og:site_name\" content=\"Software &amp; Tech Hiring Insights | BEON.tech Blog\" \/>\n<meta property=\"article:published_time\" content=\"2026-05-05T17:33:55+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2026-05-05T17:45:42+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/beon.tech\/blog\/wp-content\/uploads\/2026\/05\/ChatGPT-Image-5-may-2026-02_44_57-p.m.png\" \/>\n\t<meta property=\"og:image:width\" content=\"1672\" \/>\n\t<meta property=\"og:image:height\" content=\"941\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"author\" content=\"Matias Bustamante\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@beontechok\" \/>\n<meta name=\"twitter:site\" content=\"@beontechok\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Matias Bustamante\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"6 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\\\/\\\/beontech.wpengine.com\\\/ai-powered-querying\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/beontech.wpengine.com\\\/ai-powered-querying\\\/\"},\"author\":{\"name\":\"Matias Bustamante\",\"@id\":\"https:\\\/\\\/beon.tech\\\/blog\\\/#\\\/schema\\\/person\\\/42573ace988e25cbd44f2aaee56f5d09\"},\"headline\":\"AI-Powered Document Search with LangChain, ChatD &amp; NotebookLM\",\"datePublished\":\"2026-05-05T17:33:55+00:00\",\"dateModified\":\"2026-05-05T17:45:42+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/beontech.wpengine.com\\\/ai-powered-querying\\\/\"},\"wordCount\":1151,\"image\":{\"@id\":\"https:\\\/\\\/beontech.wpengine.com\\\/ai-powered-querying\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/beon.tech\\\/blog\\\/wp-content\\\/uploads\\\/2026\\\/05\\\/ChatGPT-Image-5-may-2026-02_44_57-p.m.png\",\"articleSection\":[\"Webinars\"],\"inLanguage\":\"en-US\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/beontech.wpengine.com\\\/ai-powered-querying\\\/\",\"url\":\"https:\\\/\\\/beontech.wpengine.com\\\/ai-powered-querying\\\/\",\"name\":\"AI-Powered Querying for PDFs | Webinar by Matias Bustamante\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/beon.tech\\\/blog\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/beontech.wpengine.com\\\/ai-powered-querying\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/beontech.wpengine.com\\\/ai-powered-querying\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/beon.tech\\\/blog\\\/wp-content\\\/uploads\\\/2026\\\/05\\\/ChatGPT-Image-5-may-2026-02_44_57-p.m.png\",\"datePublished\":\"2026-05-05T17:33:55+00:00\",\"dateModified\":\"2026-05-05T17:45:42+00:00\",\"author\":{\"@id\":\"https:\\\/\\\/beon.tech\\\/blog\\\/#\\\/schema\\\/person\\\/42573ace988e25cbd44f2aaee56f5d09\"},\"description\":\"Learn three practical ways to use AI-powered querying for PDFs: LangChain + OpenAI (Python), ChatD with local LLMs, and Google NotebookLM.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/beontech.wpengine.com\\\/ai-powered-querying\\\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/beontech.wpengine.com\\\/ai-powered-querying\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/beontech.wpengine.com\\\/ai-powered-querying\\\/#primaryimage\",\"url\":\"https:\\\/\\\/beon.tech\\\/blog\\\/wp-content\\\/uploads\\\/2026\\\/05\\\/ChatGPT-Image-5-may-2026-02_44_57-p.m.png\",\"contentUrl\":\"https:\\\/\\\/beon.tech\\\/blog\\\/wp-content\\\/uploads\\\/2026\\\/05\\\/ChatGPT-Image-5-may-2026-02_44_57-p.m.png\",\"width\":1672,\"height\":941},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/beontech.wpengine.com\\\/ai-powered-querying\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/beon.tech\\\/blog\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"AI-Powered Document Search with LangChain, ChatD &amp; NotebookLM\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/beon.tech\\\/blog\\\/#website\",\"url\":\"https:\\\/\\\/beon.tech\\\/blog\\\/\",\"name\":\"Software &amp; Tech Hiring Insights | BEON.tech Blog\",\"description\":\"\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/beon.tech\\\/blog\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/beon.tech\\\/blog\\\/#\\\/schema\\\/person\\\/42573ace988e25cbd44f2aaee56f5d09\",\"name\":\"Matias Bustamante\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/beon.tech\\\/blog\\\/wp-content\\\/uploads\\\/2026\\\/05\\\/BEON-FHD-DIA-1-329-scaled-96x96.jpg\",\"url\":\"https:\\\/\\\/beon.tech\\\/blog\\\/wp-content\\\/uploads\\\/2026\\\/05\\\/BEON-FHD-DIA-1-329-scaled-96x96.jpg\",\"contentUrl\":\"https:\\\/\\\/beon.tech\\\/blog\\\/wp-content\\\/uploads\\\/2026\\\/05\\\/BEON-FHD-DIA-1-329-scaled-96x96.jpg\",\"caption\":\"Matias Bustamante\"},\"url\":\"https:\\\/\\\/beon.tech\\\/blog\\\/author\\\/matias-bustamante\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"AI-Powered Querying for PDFs | Webinar by Matias Bustamante","description":"Learn three practical ways to use AI-powered querying for PDFs: LangChain + OpenAI (Python), ChatD with local LLMs, and Google NotebookLM.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/beon.tech\/blog\/ai-powered-querying\/","og_locale":"en_US","og_type":"article","og_title":"AI-Powered Querying for PDFs | Webinar by Matias Bustamante","og_description":"Learn three practical ways to use AI-powered querying for PDFs: LangChain + OpenAI (Python), ChatD with local LLMs, and Google NotebookLM.","og_url":"https:\/\/beon.tech\/blog\/ai-powered-querying\/","og_site_name":"Software &amp; Tech Hiring Insights | BEON.tech Blog","article_published_time":"2026-05-05T17:33:55+00:00","article_modified_time":"2026-05-05T17:45:42+00:00","og_image":[{"width":1672,"height":941,"url":"https:\/\/beon.tech\/blog\/wp-content\/uploads\/2026\/05\/ChatGPT-Image-5-may-2026-02_44_57-p.m.png","type":"image\/png"}],"author":"Matias Bustamante","twitter_card":"summary_large_image","twitter_creator":"@beontechok","twitter_site":"@beontechok","twitter_misc":{"Written by":"Matias Bustamante","Est. reading time":"6 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/beontech.wpengine.com\/ai-powered-querying\/#article","isPartOf":{"@id":"https:\/\/beontech.wpengine.com\/ai-powered-querying\/"},"author":{"name":"Matias Bustamante","@id":"https:\/\/beon.tech\/blog\/#\/schema\/person\/42573ace988e25cbd44f2aaee56f5d09"},"headline":"AI-Powered Document Search with LangChain, ChatD &amp; NotebookLM","datePublished":"2026-05-05T17:33:55+00:00","dateModified":"2026-05-05T17:45:42+00:00","mainEntityOfPage":{"@id":"https:\/\/beontech.wpengine.com\/ai-powered-querying\/"},"wordCount":1151,"image":{"@id":"https:\/\/beontech.wpengine.com\/ai-powered-querying\/#primaryimage"},"thumbnailUrl":"https:\/\/beon.tech\/blog\/wp-content\/uploads\/2026\/05\/ChatGPT-Image-5-may-2026-02_44_57-p.m.png","articleSection":["Webinars"],"inLanguage":"en-US"},{"@type":"WebPage","@id":"https:\/\/beontech.wpengine.com\/ai-powered-querying\/","url":"https:\/\/beontech.wpengine.com\/ai-powered-querying\/","name":"AI-Powered Querying for PDFs | Webinar by Matias Bustamante","isPartOf":{"@id":"https:\/\/beon.tech\/blog\/#website"},"primaryImageOfPage":{"@id":"https:\/\/beontech.wpengine.com\/ai-powered-querying\/#primaryimage"},"image":{"@id":"https:\/\/beontech.wpengine.com\/ai-powered-querying\/#primaryimage"},"thumbnailUrl":"https:\/\/beon.tech\/blog\/wp-content\/uploads\/2026\/05\/ChatGPT-Image-5-may-2026-02_44_57-p.m.png","datePublished":"2026-05-05T17:33:55+00:00","dateModified":"2026-05-05T17:45:42+00:00","author":{"@id":"https:\/\/beon.tech\/blog\/#\/schema\/person\/42573ace988e25cbd44f2aaee56f5d09"},"description":"Learn three practical ways to use AI-powered querying for PDFs: LangChain + OpenAI (Python), ChatD with local LLMs, and Google NotebookLM.","breadcrumb":{"@id":"https:\/\/beontech.wpengine.com\/ai-powered-querying\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/beontech.wpengine.com\/ai-powered-querying\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/beontech.wpengine.com\/ai-powered-querying\/#primaryimage","url":"https:\/\/beon.tech\/blog\/wp-content\/uploads\/2026\/05\/ChatGPT-Image-5-may-2026-02_44_57-p.m.png","contentUrl":"https:\/\/beon.tech\/blog\/wp-content\/uploads\/2026\/05\/ChatGPT-Image-5-may-2026-02_44_57-p.m.png","width":1672,"height":941},{"@type":"BreadcrumbList","@id":"https:\/\/beontech.wpengine.com\/ai-powered-querying\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/beon.tech\/blog\/"},{"@type":"ListItem","position":2,"name":"AI-Powered Document Search with LangChain, ChatD &amp; NotebookLM"}]},{"@type":"WebSite","@id":"https:\/\/beon.tech\/blog\/#website","url":"https:\/\/beon.tech\/blog\/","name":"Software &amp; Tech Hiring Insights | BEON.tech Blog","description":"","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/beon.tech\/blog\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Person","@id":"https:\/\/beon.tech\/blog\/#\/schema\/person\/42573ace988e25cbd44f2aaee56f5d09","name":"Matias Bustamante","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/beon.tech\/blog\/wp-content\/uploads\/2026\/05\/BEON-FHD-DIA-1-329-scaled-96x96.jpg","url":"https:\/\/beon.tech\/blog\/wp-content\/uploads\/2026\/05\/BEON-FHD-DIA-1-329-scaled-96x96.jpg","contentUrl":"https:\/\/beon.tech\/blog\/wp-content\/uploads\/2026\/05\/BEON-FHD-DIA-1-329-scaled-96x96.jpg","caption":"Matias Bustamante"},"url":"https:\/\/beon.tech\/blog\/author\/matias-bustamante\/"}]}},"featured_image_src":"https:\/\/beon.tech\/blog\/wp-content\/uploads\/2026\/05\/ChatGPT-Image-5-may-2026-02_44_57-p.m-600x400.png","featured_image_src_square":"https:\/\/beon.tech\/blog\/wp-content\/uploads\/2026\/05\/ChatGPT-Image-5-may-2026-02_44_57-p.m-600x600.png","author_info":{"display_name":"Matias Bustamante","author_link":"https:\/\/beon.tech\/blog\/author\/matias-bustamante\/"},"_links":{"self":[{"href":"https:\/\/beon.tech\/blog\/wp-json\/wp\/v2\/posts\/4458","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/beon.tech\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/beon.tech\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/beon.tech\/blog\/wp-json\/wp\/v2\/users\/52"}],"replies":[{"embeddable":true,"href":"https:\/\/beon.tech\/blog\/wp-json\/wp\/v2\/comments?post=4458"}],"version-history":[{"count":0,"href":"https:\/\/beon.tech\/blog\/wp-json\/wp\/v2\/posts\/4458\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/beon.tech\/blog\/wp-json\/wp\/v2\/media\/4461"}],"wp:attachment":[{"href":"https:\/\/beon.tech\/blog\/wp-json\/wp\/v2\/media?parent=4458"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/beon.tech\/blog\/wp-json\/wp\/v2\/categories?post=4458"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/beon.tech\/blog\/wp-json\/wp\/v2\/tags?post=4458"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}