Posts

Resumable MCP: How to Build Servers and Clients with Automatic Checkpoints

Learn to build resumable MCP servers and clients with automatic checkpoint capabilities. Complete tutorial on implementing task interruption handling, state persistence, and recovery for long-running processes in Model Control Protocol. Includes practical code with FastMCP, persistent session management, and real-world examples for processes that can be interrupted and resumed from where they left off.

Stream Information in MCP: Complete Guide to Real-time Progress Updates with FastMCP

Learn how to implement real-time streaming in MCP (Model Context Protocol) applications using FastMCP. This comprehensive guide shows you how to create MCP servers and clients that support progress updates and streaming information for long-running tasks. You'll build streaming-enabled tools that provide real-time feedback during data processing, file uploads, monitoring tasks, and other time-intensive operations. Discover how to use StreamableHttpTransport, implement progress handlers with Context, and create visual progress bars that enhance user experience when working with MCP applications that require continuous feedback.

MCP: Complete Guide to Create servers and clients MCP (Model Context Protocol) with FastMCP

Learn what is the Model Context Protocol (MCP), the open-source standard developed by Anthropic that revolutionizes how AI models interact with external tools. In this practical and detailed guide, I take you step by step in creating an MCP server and client from scratch using the fastmcp library. You will build an "intelligent" AI agent with Claude Sonnet, capable of interacting with the GitHub API to query issues and repository information. We will cover from basic concepts to advanced features like filtering tools by tags, server composition, static resources and dynamic templates (resource templates), prompt generation, and secure authentication. Discover how MCP can standardize and simplify the integration of tools in your AI applications, analogously to how USB unified peripherals!

Agents patterns

Are your agents falling short? Elevate your AI projects with advanced patterns: ReAct, planning, multi-agents, and more. Practical guide with code!

LangGraph: Revolutionize your AI agents

🚀 Revolutionize your AI agents! 🧠 LangGraph is not just another library, it's the orchestration framework that gives you total control to build complex agents, with long-term memory and even human intervention! Say goodbye to basic chatbots, it's time to create true intelligence. Dive into this post and discover it!

Create virtual environments with uv

Learn how to create virtual environments with uv, a package manager and environment for Python written in Rust, which makes it very fast. If you have had problems with the waiting times using conda, or want a faster and easier alternative to venv, enter and see how to use uv.

Make a real-time AI application with FastRTC

If you have problems making a real-time AI application, FastRTC can help you. In this post I explain how to use it.

Generate videos with Wan2.1-T2V-14B and Inference Providers

Do you want to have your own Sora, but also generate good videos? In this post I explain how to do it with HuggingFace Inference Providers and Replicate.

Deploy backend with LLM in HuggingFace

Do you want to deploy a backend with your own LLM? In this post I explain how to do it with HuggingFace Spaces, FastAPI, Langchain and Docker.

Langchain with open source integrations

Learn how to use Langchain with the most popular open-source integrations. In this post, we will explore how to integrate Langchain with ChromaDB, Ollama and HuggingFace.

unittest

Refactoriza sin miedo 🛠️ gracias a los test unitarios. En este post vas a aprender a hacer test unitarios en Python con la librería unittest 🐍✅ y a medir la cobertura de test 📊 para que puedas hacer todas las modificaciones que quieras sin miedo a romper nada 💪

RAG fundamentals

Forget about Ctrl+F! 🤯 With RAG, your documents will answer your questions directly. 😎 Step-by-step tutorial with Hugging Face and ChromaDB. Unleash the power of AI (and show off to your friends)! 💪

Conventional commits

😠 Are your commits written in alien language? 👽 Join the club! 😅 Learn Conventional Commits in Python and stop torturing your team with cryptic messages. git-changelog and commitizen will be your new best friends. 🤝

DoLa – Decoding by Contrasting Layers Improves Factuality in Large Language Models

Have you ever talked to an LLM and they answered you something that sounds like they've been drinking machine coffee all night long 😂 That's what we call a hallucination in the LLM world! But don't worry, because it's not that your language model is crazy (although it can sometimes seem that way 🤪). The truth is that LLMs can be a bit... creative when it comes to generating text. But thanks to DoLa, a method that uses contrast layers to improve the feasibility of LLMs, we can keep our language models from turning into science fiction writers 😂. In this post, I'll explain how DoLa works and show you a code example so you can better understand how to make your LLMs more reliable and less prone to making up stories. Let's save our LLMs from insanity and make them more useful! 🚀

QLoRA: Efficient Finetuning of Quantized LLMs

Hello everyone! 🤗 Today we are going to talk about QLoRA, the technique that will allow you to make your language models more efficient and faster ⏱️. But how does it do it? 🤔 Well, first it uses quantization to reduce the size of the model weights, which saves memory and speed 📈. Then, it applies LoRA (Low-Rank Adaptation), which is like a superpower that allows the model to adapt to new data without retraining from scratch 💪. And, for you to see how it works in practice, I leave you with a code example that will make you say 'Eureka!' 🎉 Let's dive into the world of QLoRA and discover how we can make our models smarter and more efficient! 🤓

GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers

Attention developers! 🚨 Do you have a language model that is too big and heavy for your application? 🤯 Don't worry, GPTQ is here to help you! 🤖 This quantization algorithm is like a wizard that makes unnecessary bits and bytes disappear, reducing the size of your model without losing too much precision. 🎩 It's like compressing a file without losing quality - it's a way to make your models more efficient and faster! 🚀

llm.int8() – 8-bit Matrix Multiplication for Transformers at Scale

Get ready to save space and speed up your models! 💥 In this post, I'm going to explore the llm.int8() method, a quantization technique that allows you to reduce the size of your machine learning models without sacrificing too much accuracy. 📊 That means you'll be able to train and deploy larger and more complex models in less space and with lower resource consumption! 💻 Let's see how to use llm.int8() with transformers to quantize a model and make it more efficient, without losing the essence of its artificial intelligence. 🤖

LLMs quantization

Imagine having a giant language model that can answer any question, from the capital of France to the perfect brownie recipe! 🍞️🇫🇷 But what happens when that model has to fit on a mobile device? 📱 That's where quantization comes in! 🎉 This technique allows us to reduce the size of models without sacrificing their accuracy, which means we can enjoy artificial intelligence on our mobile devices without the need for a supercomputer. 💻 It's like compressing an elephant into a shoebox, but without crushing the elephant! 🐘😂

LoRA – low rank adaptation of large language models

Get ready to take your model adaptation to the next level with LoRA! 🚀 This low-rank adaptation technique is like a superhero cape for your neural networks - it helps them learn new tricks without forgetting old ones 🤯. And the best thing about it? You can implement it in just a few lines of PyTorch code 💻 And if you're like me, a poor GPU guy struggling with limited resources 💸, LoRA is like a godsend: it lets you adapt your models without training them from scratch or spending a fortune on hardware 🙏 Check out the post for a step-by-step guide and a practical example!

Fine tuning Florence-2

You've already got Florence-2 on your radar, but now you want to take it to the next level! 🚀 It's time for fine-tuning! 💻 In this post, I'll walk you step-by-step through the process of adapting this multimodal model to your specific needs. 📊 From preparing your data to setting up hyperparameters, I'll show you how to get the most out of Florence-2. 💡 With PyTorch and Python, we'll make this model fit your needs and become your trusted tool for solving language and vision tasks. 📈 So get ready to take your model to the next level and make Florence-2 shine in all its glory! ✨

Fine tuning SMLs

😂 Tuning small language models? Please, you're not 'tuning', you're just desperately trying to get some sense out of a model that's about as useful as a chocolate teapot 🍫🚽. But hey, who doesn't love a good challenge? Go ahead, waste your GPU hours, and maybe, just maybe, you'll get a model that can distinguish between 'hello' and 'goodbye' 🤖. Good luck, and don't say I didn't warn you 😜

GPT-2 – Language Models are Unsupervised Multitask Learners

Unlock the power of text generation with GPT-2, the latest open model from OpenAI 💸! 🚀 In this post, I take you by the hand through the architecture behind this model, and show you how to fine-tune it 😜, code included. Read more and find out how GPT-2 can make your words more interesting than a human's 💬 (or at least, than a bored human's) 😉

Florence-2

Attention everyone! 🚨 We have a new king in town! 👑 Florence-2, the multimodal model that is revolutionizing the artificial intelligence game. 🤯 With only 200M parameters in its base version (or 700M in its large version, for those who want with everything 💥), this model is sweeping the benchmarks, beating models with 10 and 100 times more parameters. 🤯 It's like having a Swiss army knife in your AI toolkit! 🗡️ Modify the prompt and voila, Florence-2 adapts to any task you need. 🔧 In short, Florence-2 is the new SOTA (State-Of-The-Art) model in MLLMs (Multimodal Language Models) and you can't miss it. 🚀 Let's see what other surprises this beast has in store! 🤔

GPT1 – Improving Language Understanding by Generative Pre-Training

Unlock the power of language!!!! 💥 In my last post, I take you by the hand through the GPT-1 paper, explaining in a clear and concise way how this pioneer model in natural language processing works. And not only that! I also show you how to fine-tuning the model so you can adapt it to your specific needs 📊 Don't miss the opportunity to learn about one of the most influential models in history! 🚀 Read my post and find out how you can improve your artificial intelligence skills! 📄

BPE tokenizer

🔍 Discover the secret of tokenization! 🔑 I reveal you the mysteries of BPE (Byte Pair Encoding) tokenization, one of the most popular and effective methods to split text into tokens. Learn how to to tokenize with BPE! 💻 Read my post and discover the tips and tricks to master tokenization with BPE! 📄

Hugging Face Optimun

Attention, slow PyTorch models! 🐌 Optimun, the Hugging Face library, comes to the rescue to speed up your workouts and inferences. With Optimun, you can forget about speed issues and enjoy more speed and efficiency 🕒️. And best of all, it's PyTorch compatible - go on, give your models a boost with Optimun! 💻

Ollama

🏃 Run any LLM on your computer, whether it is a NASA spacecraft 🚀 or a coffee maker 🫖. With Ollama you have the facility to use any language model as if you were using docker 🐳. Moreover, as the models are quantized 📥, they will fit in memory 🧠 and they will go very fast 🚄

HuggingFace Accelerate

Speed up your training with HuggingFace Accelerate

Hugging Face Evaluate

Forget sleepless nights calculating metrics and losing your mind with evaluate for your NLP models! 😩 Hugging Face's evaluate library is the answer to your prayers, allowing you to evaluate the performance of your models easily and quickly 🚀. With evaluate, you can say goodbye to manual calculations and hello to full automation 🤗 leaving you more time to focus on what really matters 🤔: improving your models and revolutionizing the world of AI 💻

Hugging Face Datasets

Wake up, AI developer! ☕️ Stop wasting your time loading and preprocessing data like a slave 🙅‍♂️ and discover Hugging Face's datasets library, the solution to all your NLP problems (and some of your life problems, too). With datasets, you can access a collection of high-quality datasets that will make you forget you ever had to deal with CSVs and pandas 🐼 So don't lag behind, upgrade your skills and enter the 21st century of AI with datasets! 💥

Hugging Face Transformers

🤖 Transform your world with Hugging Face Transformers! 🚀 Ready to make magic with natural language? From super-fast techniques with pipeline 🌪️ to ninja tricks with AutoModel 🥷, this post takes you by the hand on an epic adventure into the NLP universe. Explore how to generate text that surprises, train models that dazzle, and share your creations on the Hugging Face Hub like a pro. Get ready to code and laugh, because the future of NLP is now and it's hilarious! 😂

git-sim

If you are afraid 😱 to use git commands, come in and learn how to test them easily and graphically with git-sim 😊

Hugging Face Tokenizers

📄 ➡️ 🔤 Explore the power of the Hugging Face Tokenizers library for natural language processing in AI. Discover how this essential tool transforms text into structured data, optimizing AI model training with practical examples and Python code. Dive into the future of NLP with our expert guide

Transformers – from top to the bottom

Discover transformers 🚀. Learn the architecture behind all the new language models. Don't ask an AI, come in and learn

ChromaDB

Discover how ChromaDB redefines vector data search and management for AI applications, with ultra-fast and accurate similarity searches in high-dimensional spaces

Regular expressions

It explores the power and precision of regular expressions, essential tools in programming that allow describing patterns in text strings for efficient search, extraction and manipulation. This post unravels from the fundamentals to advanced regex techniques, offering a practical guide with clear examples to master their use in data analysis and management. Get ready to transform your approach to text processing, unlocking new possibilities in your development projects.

CSS

CSS, or Cascading Style Sheets, is a fundamental technology in web design that allows developers and designers to style and present HTML documents in a sophisticated and efficient manner. Through CSS, we can control layout, colors, fonts, and much more, enabling the creation of rich and visually appealing user experiences. This post explores the basics of CSS, from selectors and properties to the box model, offering a detailed guide for those looking to get started or improve their web design skills. With practical examples and helpful tips, it demonstrates how CSS makes it easy to separate content from presentation, improving the accessibility and maintainability of web projects.

Mixtral-8x7B

Discover the fashion model in the world of AI

Embeddings similarity measure

Discover how similarity is measured between embeddings, the basis of the attention mechanism of transformers and RAG algorithms

Embeddings

Discover the power of embeddings

Tokens

Discover what tokens are and how words are divided into minimum units of word representation

OpenAI API

🚀 Discover the power of the OpenAI API in this post! 🌟 Learn how to install the OpenAI library ✨ and I'll guide you through the first steps to become an artificial intelligence guru. 🤖 No matter if you're a curious beginner or a coding expert looking for new adventures, this post has everything you need to get started. Get ready to explore the universe of GPT for text generation and DALL-E, image analysis, all with a touch of fun and a lot of innovation! 🎉👩‍💻 Dive into the exciting world of AI and start your journey to unlimited creativity! 🌈💻

HTML

Learn HTML, the markup language used to create web pages. Learn how to structure the content of a web page and format it with HTML.

Fail2ban

🛡️ 💻 Raise your security game! Dive into this tutorial and find out how Fail2Ban helps you stop brute force attacks in their tracks. From installation to configuration, I've got you covered. #SecurityToTheMaximum

Caffeine

🚀 Wake up, Ubuntu! ☕ With Caffeine for Linux, say goodbye to those annoying sleep modes just when you least expect it. Work marathon? Eternal download? Caffeine is here to keep your machine up and running. This cool thing integrates seamlessly with your Ubuntu environment and is a breeze to use. Find out more about how to install and use Caffeine in our post - we promise your computer won't fall asleep while you read!

Conda

Conda vs Pip - the eternal battle! Find out why Conda wears the champion belt 🥊 and how Mamba is not just a dance, but the turbo your data science projects were waiting for 🚀. Dive with us into this maze of Python packages and find your way out in style 😎 🐍. #PackagingAdo

Git

Have you rewritten that code that took you hours to do? Don't panic! With GIT, say goodbye to Oops, I deleted it. Join me on this adventure and learn how to play with versions like a programming rockstar. 🎸 💾 #GITandRoll

Whisper

👂 You don't get to gossip what the first one says? With whisper you can not only do it, but also translate it if you want. Log in and learn how

SSH

Connecting securely is essential in today's digital world. Discover how SSH allows you to establish encrypted communications between your machine and remote servers. From basic concepts to advanced tips, dive into this comprehensive tutorial and master the craft of SSH. 🔐 💼 #SSHSecure

Blip2

🤯 Can you imagine that you can add images to a gpt chat and ask questions about them? With Blip2 you can, log in and learn how

Bash scripting

Feel like an old school programmer 🐧 with this post about bash scripting.

Docker

🐳 Create as many development environments as you want with Docker. No more On my machine it works

Terminal

🟢 Feel like a real hacker 🖥 knowing everything about the terminal 💻. Get inside and start handling ⌨ the terminal like a pro.

Data management with Pandas

📊 Manage data 📝 easily with Pandas 📖

Matrix calculation with NumPy

🟢 Learn to do all the matrix calculation you want in an agile way with NumPy

Python introduction

Python 🐍 is one of the programming languages 💻 MOST USED. Enter and learn everything you need about Python 🐍

Continue reading

Resumable MCP: How to Build Servers and Clients with Automatic Checkpoints

Stream Information in MCP: Complete Guide to Real-time Progress Updates with FastMCP

MCP: Complete Guide to Create servers and clients MCP (Model Context Protocol) with FastMCP

Last posts -->

Have you seen these projects?

Horeca chatbot

Naviground

Subtify

View all projects -->

Do you want to apply AI in your project? Contact me!

Do you want to improve with these tips?

Memory profiler

See the memory usage of a script

DataLoader with pin_memory and num_workers

Increase DataLoader performance with pin_memory and num_workers

py-smi

Python library to get GPU data like `nvidia-smi`

Last tips -->

Use this locally

Hugging Face spaces allow us to run models with very simple demos, but what if the demo breaks? Or if the user deletes it? That's why I've created docker containers with some interesting spaces, to be able to use them locally, whatever happens. In fact, if you click on any project view button, it may take you to a space that doesn't work.