LLMs for Code Generation Training in Madrid, Spain

This hands-on training explores how to effectively use Large Language Models (LLMs) such as GPT, Claude, and others for generating clean, functional code and high-quality software documentation. Participants will learn how to integrate LLMs into the software development lifecycle to improve efficiency, accuracy, and maintainability.

AI FOR CODE GENERATION

1. Introduction to LLMs in Software Development

• Overview of LLMs and how they assist in coding
• Tools overview: GPT-4, Claude, GitHub Copilot, etc.
• When (and when not) to use AI in development

2. Prompt Engineering for Coding Tasks

• Writing effective prompts for different coding levels
• Breaking complex problems into promptable chunks
• Code commenting, constraints, and testing in prompts

3. Code Generation Use Cases

• Writing functions, modules, classes using LLMs
• Working across multiple programming languages
• Using AI to generate test cases and unit tests

4. Refactoring and Debugging with LLMs

• Prompting LLMs to improve existing code
• Error explanation and bug fixing workflows
• Performance optimization suggestions from AI

AUTOMATED SOFTWARE DOCUMENTATION

1. Generating Developer-Focused Documentation

• Creating inline comments, docstrings, and type hints
• Auto-generating README.md files and usage examples
• Prompting for consistent style and format

2. API Documentation and Guides

• Using LLMs to create Swagger/OpenAPI documentation
• Generating integration guides and code walkthroughs
• Multi-language support in documentation

3. Documentation from Code and Vice Versa

• Reverse engineering from code to documentation
• Auto-generating code from detailed spec documentation
• Maintaining document-code consistency with AI

4. Best Practices and Tool Integration

• Integrating AI into IDEs, repos, and CI/CD pipelines
• Validating and reviewing AI-generated documentation
• Avoiding hallucinations, bias, and incorrect code patterns

5. Final Wrap-Up

• Build a documented mini application with AI
• Showcase end-to-end workflow: code + documentation
• Review, feedback, and next steps in LLM adoption