Max.ai Analytics Assistant Release Notes - Version 24.20
Release Date: October 30th, 2024

Introduction

We’re excited to announce version 24.20 of the Max.ai Analytics Assistant SDK, introducing powerful new features to enhance your analytics experience. This release focuses on expanded query capabilities and increased flexibility in metric display, making data insights even more accessible and customizable.

Key Improvements

Enhanced SDK Query Capabilities for Chat Entries

The SDK now allows querying across the entire set of chat entries, rather than being limited to the current user or a specific thread. This feature includes options for offset, limit, and filters to refine the results, making it easier for developers and analysts to retrieve comprehensive data across user interactions.

SDK Endpoint for Gathering User Information

With a user ID, one can now query for user information such as their name through the SDK.

Prompt Library Improvements

Users can now manage versions for their prompts making it easier to confidently make changes. In addition when exporting an Assistant, any prompt library dependencies are automatically included.

Custom Metric Formatting Options

Users can now select a “Custom” option for metric formatting, which lets them input their own format string to control the appearance of metrics. For example, users can specify a format like "$,.2f" to display values as currency with two decimal places. This customization adds flexibility in presenting metrics, making data output more relevant and readable for various business contexts. With this enhancement, users can adjust the format to better suit their reporting needs and achieve more precise, context-sensitive data displays.

Questions View

Now when users go to the Questions view in Skill Studio, the question collections are automatically collapsed making it easier to navigate when you have a lot of questions and question collections.

Tested Models

This release has been tested with the following models to ensure compatibility and stability:

OpenAI:

• Chat: gpt-4o-2024-08-06
• Narrative: gpt-4o-2024-08-06
• Evaluations: gpt-4o-2024-08-06
• Embedding: text-embedding-3-small

Gemini:

• Chat: gemini-1.5-pro-002
• Narrative: gemini-1.5-pro-002
• Evaluations: gemini-1.5-pro-002
• Embedding: models/embedding-001

Anthropic:

• Chat: claude-3-5-sonnet-20241022
• Narrative: claude-3-5-sonnet-20241022
• Evaluations: claude-3-5-sonnet-20241022
• Embedding: Not Applicable