Blog Posts

Home /
Blog Posts

Transforming RAG systems with enhanced context (a langchain implementation)

Audio of the post:

I recently stumbled upon Anthropic’s fascinating post about contextual retrieval and was immediately intrigued by the potential to revolutionize RAG systems! The concept was so compelling that I decided to put it to the test with a real-world experiment using 10 years of annual reports from CUAC FM. What started as curiosity turned into a comprehensive research project. I crafted 30 carefully designed questions paired with 30 human-reviewed reference answers to rigorously evaluate whether contextual retrieval truly delivers on its promises. The results? Absolutely game-changing!

Teaching an LLM to ride the bus (Coruña edition)

I’ve been playing around with the Model Context Protocol (MCP) for a few weeks now. There’s already a lot of material out there, but I wanted to try it with a simple, real-world case: the buses in A Coruña.

Revolutionizing news with AI: How I built an automated news Podcast generator

I’ve always admired Ángel Martín’s approach to delivering news—straight to the point, no fluff, just the essentials. Inspired by that philosophy, I built an AI-powered news podcast generator focused on delivering concise, relevant news for my city, A Coruña. My goal was to create a system that keeps people informed without the need to sift through lengthy articles or multiple sources.

How I created AI-generated trivia questions

Just last week, an old teammate hit me with the question: “How can I use AI to generate random trivia questions?” At the same time, I was prepping a presentation for my colleagues at DEUS, so I thought—why not turn this into a real example? And boom! The result? An AI-powered trivia generator that effortlessly creates engaging, dynamic questions! What started as a simple inquiry became a full-blown project—challenge accepted, mission accomplished!

Guardrails for LLMs: ensuring secure and reliable AI systems for Loredo bank

The rapid evolution of Large Language Models (LLMs) has unlocked transformative applications, from content generation to automated decision-making. However, deploying LLMs in real-world systems requires robust security and reliability mechanisms. This post explores essential guardrails, the role of Pydantic as an output parser, and security concerns in agentic AI approaches.

How to build a dashboard in Azure Cloud using App Insights queries with KQL generated by LLM

Building a robust and insightful dashboard in Azure Application Insights with KQL (Kusto Query Language) allows teams to monitor and analyze their application’s performance and user behavior. This guide will walk you through creating such a dashboard with examples of key performance indicators (KPIs) and corresponding charts. I don´t know nothing about KQL but I will use an LLM to generate the queries I need.

Introducing ‘Idealisto’: Your AI Chatbot for the Spanish Real Estate Market

Navigating the Spanish real estate market just got easier with Idealisto! Whether you’re a savvy investor, a first-time buyer, or a real estate professional, this cutting-edge AI chatbot is here to help you uncover trends, get legal advice, and spot market opportunities with unprecedented precision.
Read More

Introducing AI-Powered Plant Analyzer! 🌱

I’ve developed an AI-powered plant analyzer that brings together cutting-edge tools from OpenAI Vision API and Tavily, with search capabilities to explore information on gardenia.net.

Create a podcast with zero human intervention

Have you heard the buzz about Google’s Notebook LM? This tool is a game-changer for anyone curious about leveraging AI to streamline research, supercharge content creation, and effortlessly organize insights. Whether you’re a researcher, content creator, or simply an AI enthusiast, Notebook LM brings unparalleled structure and depth to your data handling.

How I fixed my coffee machine using a RAG System

When my coffee machine decided to quit on me, going through the manual was just painful and honestly, a waste of time. So, instead of giving up, I tried something different: I used a Retrieval-Augmented Generation (RAG) system with a Large Language Model (LLM) to figure it out.