Skip to main content

Advanced Usage

This guide covers advanced techniques and best practices for optimizing your modernization projects using the Smart Engineering module.

Optimizing AI Model Selection​

Different tasks in the modernization process have varying computational requirements. By strategically selecting AI models based on task complexity, you can optimize both performance and cost.

Model Selection Best Practices
  • Lightweight Models (e.g., Claude Haiku, Gemini Flash): Ideal for enrichment tasks, code indexing, and simple documentation generation
  • Robust Models (e.g., Claude Sonnet, Gemini Pro): Best for complex tasks like diagram generation, architecture analysis, and comprehensive overviews

Reindexing and Updating Projects​

As your codebase evolves, you may need to update your modernization project with new or modified files. The Smart Engineering module provides intelligent reindexing capabilities.

Automatic Change Detection​

When you reupload source code to an existing project, the system automatically:

  1. Detects New Files: Identifies files that weren't present in the previous upload
  2. Identifies Changes: Recognizes modifications to existing files
  3. Prevents Duplication: Ensures files aren't duplicated in the index
  4. Updates Metadata: Refreshes file metadata and relationships
Smart Reindexing

The reindexing process is intelligent and incremental. Only new or modified files are processed, making updates efficient even for large codebases.

Preserve Task History

Reindexing doesn't delete previous task results. You can compare old and new documentation to track how your system has evolved.