Learn the Basics

Most AI workflows involve working with data, defining agents, optimizing agent parameters, and deploying trained agents. This guide introduces you to a complete workflow implemented in Afnio, with links to learn more about each concept.

We’ll use a real-world sentiment classification task—automatically detecting positive, neutral, or negative feedback in enterprise support emails—to demonstrate how to create variables, build agents, manage datasets, and run optimization loops using Afnio’s modular API.

This guide assumes you are comfortable with Python and have a basic understanding of AI or machine learning concepts.

How to Use this Guide

If you’re familiar with deep learning frameworks like PyTorch, check out the 0. Quickstart first to quickly familiarize yourself with Afnio’s API.

If you’re new to deep learning frameworks, head right into the first section of our step-by-step guide: 1. Login to Afnio Backend.

• 0. QuickStart
• 1. Login to Afnio Backend
• 2. Variables
• 3. Datasets and DataLoaders
• 4. Build the Agent or Workflow
• 5. Automatic Differentiation
• 6. Optimization Loop
• 7. Trainer
• 8. Save, Load and Use Agent
• 9. Runs and Experiments