Python Brasil 2025

Unlocking the Power of LLMs: Creating RAG Applications
23/10/2025 , Impacta - Sala 206

In this talk, we'll explore the power of Retrieval-Augmented Generation (RAG) models, which combine the strengths of LLMs with the flexibility of retrieval-based approaches.


Large Language Models (LLMs) have revolutionized the field of natural language processing, but their true potential lies in their ability to be integrated into real-world applications. In this talk, we'll explore the power of Retrieval-Augmented Generation (RAG) models, which combine the strengths of LLMs with the flexibility of retrieval-based approaches. We'll delve into the design and implementation of RAG models, discussing the benefits and challenges of using pre-trained LLMs as a foundation. Through a hands-on example, we'll demonstrate how to build a RAG model that can be applied to a variety of real-world problems, from text generation to question-answering. By the end of this talk, attendees will gain a deeper understanding of how to unlock the full potential of LLMs and create practical, effective RAG models that drive business value.


Quais conhecimentos prévios são necessários para que seja possível acompanhar bem a sua atividade?
  • basics of docker: how to create yaml files and use docker for local deployment.
  • web development: intermediate understanding of web API development, some previous experience is a nice to have.
  • basic understanding of LLM and prompting: being familiar on basic LLM inner working and how to build prompts.
O que as pessoas que participarem podem esperar aprender na sua atividade?

From this talk a participant can expect to learn how to deploy local models, use model orchestration libraries, build agents for specific task, apply RAG architectural style and to build an entire application for a small use case.

Escolha uma ou mais áreas em que essa proposta se encaixa

Machine Learning and Artificial Intelligence, Web Development, Software Architecture

David Camacho is a transformative technology leader with over 10 years of experience in software engineering and 8 years in different data roles, driving innovation and growth in multiple industries.