> ## Documentation Index
> Fetch the complete documentation index at: https://docs.usethunder.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Introduction

> Thunder is an analytics platform for AI-powered conversations. It helps you understand what your users are talking about, identify trends, and improve your AI applications.

## What Thunder Does

Thunder ingests your conversation data and provides insights through:

* **Topic Detection** - Automatically categorizes conversations into topics
* **Satisfaction Signals** - Detects user satisfaction (CSAT) and dissatisfaction (DSAT) signals
* **Gap Detection** - Identifies knowledge gaps and missing tool capabilities
* **Repeated Requests** - Finds patterns where users have to ask the same thing multiple times
* **Usage Metrics** - Tracks sessions, users, and message volumes over time

## Core Concepts

### Instances

An instance represents a distinct product or deployment within your AI application. For example, if you're building AI clones or agents, each clone would be its own instance. If your product is a single conversational experience, you can use the default instance. All analytics in Thunder are scoped to instances.

### Sessions

A session is a single conversation between your AI and a user. Sessions belong to instances and contain one or more messages. Each session is analyzed to extract topics, detect satisfaction signals, identify gaps, and find repeated requests.

### End Users

End users are the people interacting with your AI. Thunder tracks user activity across sessions to help you understand engagement patterns and identify users who may be having trouble.

### Topics

Thunder analyzes message content to identify topics - the subjects your users are discussing. Topics help you understand what users care about and where to focus improvements.

### Satisfaction Signals (CSAT/DSAT)

Thunder automatically detects signals of user satisfaction (CSAT) and dissatisfaction (DSAT) within conversations. These signals help you measure the quality of your AI's responses and identify areas for improvement.

### Gaps

Gaps represent missing capabilities in your AI. Thunder identifies two types:

* **Knowledge Gaps** - Information the AI doesn't know but should
* **Tool Gaps** - Actions the AI can't perform but users expect

### Repeated Requests

When users have to ask the same question multiple times within a session, it often indicates the AI failed to understand or adequately respond. Thunder detects these patterns to help you improve response quality.

## Getting Started

<CardGroup cols={2}>
  <Card title="Quickstart" icon="rocket" href="/quickstart">
    Get up and running in 5 minutes
  </Card>

  <Card title="Authentication" icon="key" href="/authentication">
    Learn how API keys work
  </Card>

  <Card title="Ingest API" icon="upload" href="/api-reference/ingest/post-ingest">
    Start sending conversation data
  </Card>

  <Card title="Query API" icon="chart-bar" href="/api-reference/query/get-query">
    Retrieve analytics and metrics
  </Card>
</CardGroup>