# Next-Gen Autolog

### Overview

Next-Gen Autolog takes documentation to the next level by integrating metadata analysis with code analysis, providing richer, more accurate, and context-aware insights. Powered by [Ask AI](/25.4/introduction/readme/askai.md), it captures not just what the code does, but how it performs in real-world scenarios—enhancing understanding, usability, and maintainability.

#### The power of Metadata and Code

By integrating metadata and code analysis, Next-Gen Autolog provides documentation which is:

✅ Context-aware – Explains what the code does and how well it works.\
✅ Insight-driven – Highlights model performance, dataset biases, and risks.\
✅ Actionable – Surfaces key insights to improve model trust and usability.

### Key capabilities

#### 1. Automated context extraction with GenAI-powered Autolog

Enrich captured metadata with automated contextual information extraction, improving documentation quality and usability.

* Structured asset organization: assets are systematically categorized into sections.
* Contextual insights: automatically integrates asset creation details and relevant insights.
* Automated lineage tracking: derives relationships between assets without manual input.
* Feature engineering logic extraction: captures and documents transformation processes applied to data.

#### 2. Enriched documentation

Leverage extracted metadata to create highly detailed and nuanced documentation.

* AI-Powered understanding: Ask AI uses extracted insights to interpret and explain code functionality.
* Accurate feature engineering documentation: Captures complex data transformations and feature creation steps.
* Densely populated lineage information: Provides a comprehensive view of data dependencies and transformations.

#### 3. Automated report generation using templates and macros

With enriched metadata, reports can be generated effortlessly using predefined templates and macros.

* Advanced prompting: Ask AI applies metadata-driven prompts to generate well-structured documentation.
* Comprehensive long-form reports: Create detailed and complex documents using pre-defined macros and widgets.

### How It Works

Use[ Autolog](/25.4/introduction/readme/autolog.md) to capture all metadata from your notebook. Once the metadata is logged into Vectice:

1. Navigate to the **Iteration Page.**
2. Click on **Organize with Ask AI**.
3. **Ask AI** will:

* Reorganize iteration content
* Add structured sections, descriptions, and notes
* Build lineage between assets
* Render dataset transformations

The results are immediately visible in the iteration page and can be integrated into the documentation using Macros, Widgets or pre-defined Prompts.

{% embed url="<https://files.gitbook.com/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FbO7GsO4mI4pjZ7XjnzBT%2Fuploads%2FVKLkeBYXfGshc4Fs1EUN%2FDesign%20sans%20titre.mp4?alt=media&token=31bf7f4b-1acc-413c-8959-4d0b8093eaa4>" %}


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.vectice.com/25.4/introduction/readme/next-gen-autolog-beta.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.