{"id":16757,"date":"2024-05-09T14:48:50","date_gmt":"2024-05-09T12:48:50","guid":{"rendered":"https:\/\/alicebiometrics.com\/?p=16757"},"modified":"2024-05-22T15:11:43","modified_gmt":"2024-05-22T13:11:43","slug":"chatting-with-your-documentation-llms-rag","status":"publish","type":"post","link":"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/","title":{"rendered":"Chatting with Your Documentation: LLMs and RAG for Enhanced Information Retrieval"},"content":{"rendered":"\n

Introduction<\/h2>\n\n\n\n

Have you ever felt that you know some useful information is somewhere in your documentation, but you cannot seem to find it? Yeah, me too. It is so frustrating, but after reading this article you are going to have some notion of how to build a system that enables you and your team to find that information, faster and more comfortable than ever before.<\/p>\n\n\n\n

In Alice Biometrics, we are constantly seeking innovative ways to improve information retrieval and knowledge management. One exciting approach is combining Large Language Models (LLMs) with Retrieval-Augmented Generation (RAG) to create an interactive, chat-like experience with your documentation. In this blog post, we’ll explore how you can leverage LLMs and RAG to chat with your documentation and revolutionize the way you access and consume information, like we have done in Alice Biometrics.<\/p>\n\n\n\n

What are LLMs and RAG?<\/h2>\n\n\n\n

Large Language Models (LLMs)<\/h3>\n\n\n\n

Nowadays, many people are familiar with LLMs, but in short, these are deep learning models that have been trained on massive amounts of text data (all the internet in some cases). They can understand natural language, generate coherent responses, and perform a wide range of language-related tasks, such as question answering, summarization, and translation.<\/p>\n\n\n\n

Retrieval-Augmented Generation (RAG)<\/h3>\n\n\n\n

RAG is a framework that combines the strengths of LLMs and information retrieval techniques. It enables the model to generate answers by retrieving relevant information from an external knowledge source, such as a document corpus, and incorporating it into the generated response.<\/p>\n\n\n\n

Prepare your Knowledge Base<\/h2>\n\n\n\n

To get started, you’ll need to convert your documentation into a structured format that can be easily indexed and searched. In our case, we chose to use LangChain<\/a> as our main tool to carry out this task. The following steps describe how to build a robust and useful knowledge base, represented by a vector database, that will allow you to retrieve useful context to help the LLM to answer correctly to the user\u2019s prompt.<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

Select your Documentation<\/h3>\n\n\n\n

Identify the documents<\/strong> that are most frequently used and contain valuable information relevant to your team\u2019s work. Prioritize documents that are up-to-date and cover a wide range of topics within your domain.<\/p>\n\n\n\n

Prepare Your Documents<\/h3>\n\n\n\n

Convert all your documents<\/strong> into a uniform format, such as plain text. This process can be tedious as some of the texts from different sources (e.g. websites) may need extensive preprocess. Ensure that the text is clean, with no formatting issues or errors that could affect the retrieval process. This task is very important, specially to avoid creating noisy context or to complicate following steps in the process. As always, LangChain offers specialized Document<\/a> Loaders<\/a> that will help you to do this task more easily.<\/p>\n\n\n\n

Split your Documents in Chunks<\/h3>\n\n\n\n

Once you have collected and processed your documents, you may want to divide your documents<\/strong> into smaller, manageable chunks that can fit into your model\u2019s context window<\/a>, or simply to reduce the cost of sending these chunks as context, as we will see in the following sections. Each chunk should be a self-contained piece of information that can stand alone. This makes it easier for the RAG system to retrieve and use specific information.<\/p>\n\n\n\n

Different kinds of documents require different ways of splitting their information. For example, it is not the same splitting a markdown document than a python script, LangChain provides useful text splitters<\/a> that suit most needs, but you can always customize your own. The better your splits, the better your context will be to help answer the future questions of your teammates or users, so this step is crucial for the quality of the RAG system.<\/p>\n\n\n\n

Extract Embeddings<\/h3>\n\n\n\n

Use a pre-trained text embedding language model<\/strong> to generate embeddings for each document chunk. These embedding models are language models specially trained to produce rich numerical representations and capture semantic meaning of the text. Extracting these numerical representations allow us to efficiently retrieve similar texts. The closer the embeddings are, the more similar the topic they talk about will be, and the better context they will provide.<\/p>\n\n\n\n

These embedding models are constantly being developed and improved. In order to choose the best model, you can use some of the common leaderboards<\/a> to find the one that better suits your requirements. LangChain offers an extensive catalogue of integrations<\/a> of embedding models<\/a> for most providers and models.<\/p>\n\n\n\n

Vector Database<\/h3>\n\n\n\n

Create a vector database<\/strong>\u00a0to store the embeddings. This database will allow you to perform efficient similarity searches to find the most relevant document chunks during the retrieval phase. Currently, with the growing popularity of LLMs, there are many vector databases available<\/a>. The key metrics you want to look for in a vector database are how cheap, how accurate and how fast they are searching the most similar embeddings, specially if the amount of chunks needed for your solution is large.<\/p>\n\n\n\n

Chat With Your Data<\/h2>\n\n\n\n

Now, we go with the fun part, creating the chat and using the RAG system for retrieving context, improving the answer of the LLM.<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

Choosing the LLM<\/h3>\n\n\n\n

This will be the model that will actually chat with the user of the system. When selecting a Large Language Model for the chat of your RAG system, consider factors such as the model\u2019s size, training data, and performance on language tasks relevant to your domain, as well as the pricing. Popular models include OpenAI\u2019s GPT-3 and GPT-4, Google\u2019s Gemini, and Meta Llama2 and the recent release of Llama3, and many more. Evaluate the models based on their ability to understand and generate contextually relevant responses. LangChain offers great compatibility with most providers<\/a>.<\/p>\n\n\n\n

Retrieve Context<\/h3>\n\n\n\n

The retrieval phase is crucial in a RAG system. When the user introduces a prompt, the embedding model will produce an embedding from it. This embedding will be used to query the vector database to find the most relevant document chunks. The system uses the embeddings to perform a similarity search and retrieve the top results that provide context for the LLM to generate its response.<\/p>\n\n\n\n

Answer Citing the Documentation Retrieved<\/h3>\n\n\n\n

Once the relevant context is retrieved, a more complex prompt is built with a \u201cprompt template\u201d like the one shown below. This template instructs the model to answer the users query with information retrieved from the documentation. Furthermore, the system can be designed to cite the specific documents or sections it used, providing transparency, and allowing users to delve deeper into the topic if needed. And more importantly, hopefully finding the piece of documentation that you could never find!<\/p>\n\n\n\n

\"\"\"You are  an AI assistant for Alice Biometrics. Your mission is to help \nAlice coworkers to improve their experience and help them. \nUse the following pieces of context to answer the question at the end.\n\n{context}\n\nQuestion: {question}\nHelpful Answer: \"\"\"<\/code><\/pre>\n\n\n\n
Chat User Interface<\/h3>\n\n\n\n
Last but not least, we need to make it look nice and beautiful, and give our users an easy way to interact with this system. Streamlit<\/a> is an open-source Python framework that allows engineers to rapidly build and share beautiful, interactive web applications. In Alice Biometrics, where we have a pythonic culture, we normally use Streamlit<\/a> to create internal tools for diverse purposes, and we choose this framework for its simplicity<\/a> and ease of use for our UI. However, you might also consider some<\/a> solution<\/a> based<\/a> on<\/a> Gradio<\/a>. Both platforms allow for the quick deployment of machine learning models and can be integrated with your RAG system to provide a seamless user experience, you can see an example of our own on the image below \ud83d\ude80<\/p>\n\n\n\n
\"\"<\/figure>\n\n\n\n
Conclusion<\/h2>\n\n\n\n
By combining the power of Large Language Models and Retrieval-Augmented Generation, you can create an interactive, chat-like experience with your documentation like we did in Alice Biometrics. This approach not only improves information access and understanding, but also provides a personalized experience for your team.<\/p>\n\n\n\n
Useful Resources<\/h2>\n\n\n\n
    \n
  1. DeepLearning.AI<\/a>: at the moment they are offering an extensive catalogue of short free courses<\/a> that are really useful to build your own system, and especially to make it work. I recommend starting with LangChain: Chat with Your Data<\/a>,<\/strong> it is a nice overview of how to build your own system using LangChain, and from there you have many other courses that will help you improve your system.<\/li>\n\n\n\n
  2. LangChain Cookbook<\/a> of useful examples, both simple and more advanced<\/li>\n<\/ol>\n","protected":false},"excerpt":{"rendered":"
    Introduction Have you ever felt that you know some useful information is somewhere in your documentation, but you cannot seem to find it? Yeah, me too. It is so frustrating, but after reading this article you are going to have some notion of how to build a system that enables you and your team to […]<\/p>\n","protected":false},"author":14,"featured_media":16776,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"inline_featured_image":false,"footnotes":""},"categories":[5,62],"tags":[],"class_list":["post-16757","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-identity-verification","category-technology"],"acf":[],"yoast_head":"\nChatting with Your Documentation: LLMs and RAG<\/title>\n<meta name=\"description\" content=\"Some useful information is somewhere in your documentation, but you cannot seem to find it? Chatting with Your Documentation: LLMs and RAG\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Chatting with Your Documentation: LLMs and RAG\" \/>\n<meta property=\"og:description\" content=\"Some useful information is somewhere in your documentation, but you cannot seem to find it? Chatting with Your Documentation: LLMs and RAG\" \/>\n<meta property=\"og:url\" content=\"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/\" \/>\n<meta property=\"og:site_name\" content=\"Alice Biometrics\" \/>\n<meta property=\"article:published_time\" content=\"2024-05-09T12:48:50+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2024-05-22T13:11:43+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/alicebiometrics.com\/wp-content\/uploads\/2024\/04\/Tech-posts-Chatting-with-Your-Documentation_-LLMs-and-RAG-for-Enhanced-Information-Retrieval.png\" \/>\n\t<meta property=\"og:image:width\" content=\"1280\" \/>\n\t<meta property=\"og:image:height\" content=\"720\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"author\" content=\"Telo\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@AliceBiometrics\" \/>\n<meta name=\"twitter:site\" content=\"@AliceBiometrics\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Telo\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"7 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/\"},\"author\":{\"name\":\"Telo\",\"@id\":\"https:\/\/alicebiometrics.com\/en\/#\/schema\/person\/e892ec6adb24da3fa7398a859e48b98a\"},\"headline\":\"Chatting with Your Documentation: LLMs and RAG for Enhanced Information Retrieval\",\"datePublished\":\"2024-05-09T12:48:50+00:00\",\"dateModified\":\"2024-05-22T13:11:43+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/\"},\"wordCount\":1353,\"publisher\":{\"@id\":\"https:\/\/alicebiometrics.com\/en\/#organization\"},\"image\":{\"@id\":\"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/alicebiometrics.com\/wp-content\/uploads\/2024\/04\/Tech-posts-Chatting-with-Your-Documentation_-LLMs-and-RAG-for-Enhanced-Information-Retrieval.png\",\"articleSection\":[\"Identity Verification\",\"Technology\"],\"inLanguage\":\"en-US\"},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/\",\"url\":\"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/\",\"name\":\"Chatting with Your Documentation: LLMs and RAG\",\"isPartOf\":{\"@id\":\"https:\/\/alicebiometrics.com\/en\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/alicebiometrics.com\/wp-content\/uploads\/2024\/04\/Tech-posts-Chatting-with-Your-Documentation_-LLMs-and-RAG-for-Enhanced-Information-Retrieval.png\",\"datePublished\":\"2024-05-09T12:48:50+00:00\",\"dateModified\":\"2024-05-22T13:11:43+00:00\",\"description\":\"Some useful information is somewhere in your documentation, but you cannot seem to find it? Chatting with Your Documentation: LLMs and RAG\",\"breadcrumb\":{\"@id\":\"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/#primaryimage\",\"url\":\"https:\/\/alicebiometrics.com\/wp-content\/uploads\/2024\/04\/Tech-posts-Chatting-with-Your-Documentation_-LLMs-and-RAG-for-Enhanced-Information-Retrieval.png\",\"contentUrl\":\"https:\/\/alicebiometrics.com\/wp-content\/uploads\/2024\/04\/Tech-posts-Chatting-with-Your-Documentation_-LLMs-and-RAG-for-Enhanced-Information-Retrieval.png\",\"width\":1280,\"height\":720},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Portada\",\"item\":\"https:\/\/alicebiometrics.com\/en\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Chatting with Your Documentation: LLMs and RAG for Enhanced Information Retrieval\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/alicebiometrics.com\/en\/#website\",\"url\":\"https:\/\/alicebiometrics.com\/en\/\",\"name\":\"Alice Biometrics\",\"description\":\"Identity verification, in a second.\",\"publisher\":{\"@id\":\"https:\/\/alicebiometrics.com\/en\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/alicebiometrics.com\/en\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/alicebiometrics.com\/en\/#organization\",\"name\":\"Alice Biometrics\",\"url\":\"https:\/\/alicebiometrics.com\/en\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/alicebiometrics.com\/en\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/alicebiometrics.com\/wp-content\/uploads\/2021\/08\/h_paraClaros.svg\",\"contentUrl\":\"https:\/\/alicebiometrics.com\/wp-content\/uploads\/2021\/08\/h_paraClaros.svg\",\"width\":\"1024\",\"height\":\"1024\",\"caption\":\"Alice Biometrics\"},\"image\":{\"@id\":\"https:\/\/alicebiometrics.com\/en\/#\/schema\/logo\/image\/\"},\"sameAs\":[\"https:\/\/x.com\/AliceBiometrics\",\"https:\/\/www.linkedin.com\/company\/alicebiometrics\",\"https:\/\/www.youtube.com\/channel\/UCNB7tu9fdtLYCBoZgfxtT2w\"]},{\"@type\":\"Person\",\"@id\":\"https:\/\/alicebiometrics.com\/en\/#\/schema\/person\/e892ec6adb24da3fa7398a859e48b98a\",\"name\":\"Telo\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Chatting with Your Documentation: LLMs and RAG","description":"Some useful information is somewhere in your documentation, but you cannot seem to find it? Chatting with Your Documentation: LLMs and RAG","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:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/","og_locale":"en_US","og_type":"article","og_title":"Chatting with Your Documentation: LLMs and RAG","og_description":"Some useful information is somewhere in your documentation, but you cannot seem to find it? Chatting with Your Documentation: LLMs and RAG","og_url":"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/","og_site_name":"Alice Biometrics","article_published_time":"2024-05-09T12:48:50+00:00","article_modified_time":"2024-05-22T13:11:43+00:00","og_image":[{"width":1280,"height":720,"url":"https:\/\/alicebiometrics.com\/wp-content\/uploads\/2024\/04\/Tech-posts-Chatting-with-Your-Documentation_-LLMs-and-RAG-for-Enhanced-Information-Retrieval.png","type":"image\/png"}],"author":"Telo","twitter_card":"summary_large_image","twitter_creator":"@AliceBiometrics","twitter_site":"@AliceBiometrics","twitter_misc":{"Written by":"Telo","Est. reading time":"7 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/#article","isPartOf":{"@id":"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/"},"author":{"name":"Telo","@id":"https:\/\/alicebiometrics.com\/en\/#\/schema\/person\/e892ec6adb24da3fa7398a859e48b98a"},"headline":"Chatting with Your Documentation: LLMs and RAG for Enhanced Information Retrieval","datePublished":"2024-05-09T12:48:50+00:00","dateModified":"2024-05-22T13:11:43+00:00","mainEntityOfPage":{"@id":"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/"},"wordCount":1353,"publisher":{"@id":"https:\/\/alicebiometrics.com\/en\/#organization"},"image":{"@id":"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/#primaryimage"},"thumbnailUrl":"https:\/\/alicebiometrics.com\/wp-content\/uploads\/2024\/04\/Tech-posts-Chatting-with-Your-Documentation_-LLMs-and-RAG-for-Enhanced-Information-Retrieval.png","articleSection":["Identity Verification","Technology"],"inLanguage":"en-US"},{"@type":"WebPage","@id":"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/","url":"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/","name":"Chatting with Your Documentation: LLMs and RAG","isPartOf":{"@id":"https:\/\/alicebiometrics.com\/en\/#website"},"primaryImageOfPage":{"@id":"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/#primaryimage"},"image":{"@id":"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/#primaryimage"},"thumbnailUrl":"https:\/\/alicebiometrics.com\/wp-content\/uploads\/2024\/04\/Tech-posts-Chatting-with-Your-Documentation_-LLMs-and-RAG-for-Enhanced-Information-Retrieval.png","datePublished":"2024-05-09T12:48:50+00:00","dateModified":"2024-05-22T13:11:43+00:00","description":"Some useful information is somewhere in your documentation, but you cannot seem to find it? Chatting with Your Documentation: LLMs and RAG","breadcrumb":{"@id":"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/#primaryimage","url":"https:\/\/alicebiometrics.com\/wp-content\/uploads\/2024\/04\/Tech-posts-Chatting-with-Your-Documentation_-LLMs-and-RAG-for-Enhanced-Information-Retrieval.png","contentUrl":"https:\/\/alicebiometrics.com\/wp-content\/uploads\/2024\/04\/Tech-posts-Chatting-with-Your-Documentation_-LLMs-and-RAG-for-Enhanced-Information-Retrieval.png","width":1280,"height":720},{"@type":"BreadcrumbList","@id":"https:\/\/alicebiometrics.com\/en\/chatting-with-your-documentation-llms-rag\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Portada","item":"https:\/\/alicebiometrics.com\/en\/"},{"@type":"ListItem","position":2,"name":"Chatting with Your Documentation: LLMs and RAG for Enhanced Information Retrieval"}]},{"@type":"WebSite","@id":"https:\/\/alicebiometrics.com\/en\/#website","url":"https:\/\/alicebiometrics.com\/en\/","name":"Alice Biometrics","description":"Identity verification, in a second.","publisher":{"@id":"https:\/\/alicebiometrics.com\/en\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/alicebiometrics.com\/en\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/alicebiometrics.com\/en\/#organization","name":"Alice Biometrics","url":"https:\/\/alicebiometrics.com\/en\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/alicebiometrics.com\/en\/#\/schema\/logo\/image\/","url":"https:\/\/alicebiometrics.com\/wp-content\/uploads\/2021\/08\/h_paraClaros.svg","contentUrl":"https:\/\/alicebiometrics.com\/wp-content\/uploads\/2021\/08\/h_paraClaros.svg","width":"1024","height":"1024","caption":"Alice Biometrics"},"image":{"@id":"https:\/\/alicebiometrics.com\/en\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/x.com\/AliceBiometrics","https:\/\/www.linkedin.com\/company\/alicebiometrics","https:\/\/www.youtube.com\/channel\/UCNB7tu9fdtLYCBoZgfxtT2w"]},{"@type":"Person","@id":"https:\/\/alicebiometrics.com\/en\/#\/schema\/person\/e892ec6adb24da3fa7398a859e48b98a","name":"Telo"}]}},"_links":{"self":[{"href":"https:\/\/alicebiometrics.com\/en\/wp-json\/wp\/v2\/posts\/16757","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/alicebiometrics.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/alicebiometrics.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/alicebiometrics.com\/en\/wp-json\/wp\/v2\/users\/14"}],"replies":[{"embeddable":true,"href":"https:\/\/alicebiometrics.com\/en\/wp-json\/wp\/v2\/comments?post=16757"}],"version-history":[{"count":2,"href":"https:\/\/alicebiometrics.com\/en\/wp-json\/wp\/v2\/posts\/16757\/revisions"}],"predecessor-version":[{"id":16775,"href":"https:\/\/alicebiometrics.com\/en\/wp-json\/wp\/v2\/posts\/16757\/revisions\/16775"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/alicebiometrics.com\/en\/wp-json\/wp\/v2\/media\/16776"}],"wp:attachment":[{"href":"https:\/\/alicebiometrics.com\/en\/wp-json\/wp\/v2\/media?parent=16757"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/alicebiometrics.com\/en\/wp-json\/wp\/v2\/categories?post=16757"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/alicebiometrics.com\/en\/wp-json\/wp\/v2\/tags?post=16757"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}