Ai on Rafael Zago

How I use AI in development (Cursor and Claude)

Wed, 22 Jul 2026 00:00:00 -0300

The question I hear most about day-to-day AI is not “which model do you use?” It is “how do you organize this so it does not turn into a mess?”

I use AI constantly, in code and on this blog. What changed my routine was not a miracle prompt. It was treating the agent like someone who joins the project with no context: if the rule is not in a file, it does not exist. Skills, AGENTS.md, agents with narrow roles, and me reviewing the result. No magic.

Claude Code: from PRD to pull request with agents

Fri, 08 May 2026 00:00:00 -0300

Introduction

I discovered Claude Code a few months ago when our team was struggling to keep code reviews consistent. After six weeks with the tool, we cut average PR time from 4 days to 1 day — and that was only the start.

Claude Code automates a large part of the development workflow. PRDs, TDD, code review, even MCP setup. It is not magic, but it works well enough to change the day-to-day.

How an AI works when you talk to it

Wed, 24 Dec 2025 00:00:00 -0300

Unpacking large language models

When you type a message to ChatGPT or Claude, what happens on the other side is simpler than it looks—but it runs at a scale that is hard to picture.

The essence: a digital football commentator

Picture this: World Cup final, Brazil vs Argentina, last minutes of the match. Galvão Bueno is calling the game, the ball reaches the Brazilian striker, he beats one defender, then another… and suddenly the audio cuts out. The broadcast keeps rolling, but the commentator’s voice is gone.

Skupper + InstructLab: controlling and protecting AI (act 3)

Mon, 05 May 2025 00:00:00 -0300

See the Solution in Action

In this article we deploy the InstructLab chatbot on Kubernetes step by step, using Skupper to securely connect a private AI model to a public interface. This is the hands-on continuation of the solution pattern from the previous article.

Concepts and Commands Used in the Demo

NOTE: The following commands configure the environment and deploy the InstructLab chatbot. They are taken from the InstructLab project and adapted for this demo.

Skupper + InstructLab: controlling and protecting AI (act 2)

Sun, 04 May 2025 00:00:00 -0300

See the Solution in Action

In this article we prepare the full environment to serve the generative DeepSeek AI model with InstructLab: download the model, convert it to GGUF, and deploy the InstructLab chatbot. We also expose the service securely with Skupper.

Concepts and Commands Used in the Demo

NOTE Make sure you explain what each Skupper command does the first time you use it, especially for people new to Skupper. The following commands should be explained:

Skupper + InstructLab: controlling and protecting AI (act 1)

Fri, 02 May 2025 00:00:00 -0300

The story behind this solution pattern

Growing demand for AI-driven applications brings a hard problem: how do you deploy and operate AI models securely in environments that need strict data protection, while those models still have to be reachable by public services? That need became clear while building a local AI chatbot meant to handle sensitive and proprietary information — we needed a design that kept the model inside a protected environment.

InstructLab and Skupper: local AI without exposing data

Thu, 01 Aug 2024 00:00:00 -0300

Welcome to the Ollama Pilot.

Problem to solve

The main goal of this project is to create a secure connection between two sites, enabling the communication between the engineer machine and an Instruct Lab Model. The merlinite-7b-lab-Q4_K_M.gguf model will be used for the chatbot, and it is available in the Instruct Lab. The license of the model is available in the Instruct Labs.

But, why the banner? Well, the engineer needs to know who is better, Lebron or Jordan. The chatbot will be responsible for answering this question. The chatbot will receive the user input and send it to the llama3 model. The response from the merlinite model will be sent back to the user.

OpenShift AI and Skupper: insurance fraud prevention

Mon, 17 Jun 2024 00:00:00 -0300

Description

This workshop demonstrates how to use Skupper to connect local data services to cloud-based AI/ML environments. The workshop includes a Go application in a podman container that exposes internal data for Skupper connection. The AI/ML model training is performed in an OpenShift AI cluster on AWS using Openshift AI/ML services.

Disclaimer

This lab uses the example from the AI/ML Workshop created by the Red Hat AI Services team. The original workshop is available on GitHub and includes all the necessary information to run the lab. The lab was adapted to use Skupper to connect the local data services to the cloud-based AI/ML environment.