> ## Documentation Index
> Fetch the complete documentation index at: https://cortex-e852fafe-t3code-rewrite-docs-declutter.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Introduction

> Store documents and user memories, find relevant information, and use it in your AI application.

## What HydraDB is

HydraDB stores documents and user memories so your AI application can find the information it needs for an answer. That information is the model's **context**: the material you include alongside a user's question.

* **Knowledge:** Source material such as product docs, support tickets, Slack messages, and internal policies.
* **Memories:** Information to carry between conversations, such as a user's preferences, previous decisions, or support history.

Your application sends content to HydraDB, searches it when needed, and passes the results to the AI model that writes the answer.

## The problem we're solving

Searching by meaning helps an agent find relevant documents, but the same question can need different information for different people. A sales representative querying "project Acme" needs the latest sales deck and competitive notes. An engineer running the same query needs the changelog and architecture decisions.

HydraDB lets your application choose which data to search and provide hints about the task. It combines search by meaning, keyword matching, and relationships between your content behind one API. Your application specifies which collections to search and sends the caller's [permissions](/essentials/v2/access-control); HydraDB does not know who is asking unless you tell it.

**Skip ahead:** [Quickstart](/get-started/v2/quickstart) · [API Reference](/api-reference/v2) · [SDKs](/api-reference/v2/sdks)

***

## How it fits into your application

1. **Store content.** Upload documents, save memories, or use a connector to sync an app such as Slack or Google Drive.
2. **Wait for indexing.** HydraDB processes the content in the background to make it searchable.
3. **Search.** Send a question and choose the database and collections to search. A database separates customers' data; collections organize it by user, team, or project.
4. **Use the results.** HydraDB returns relevant passages and source details. Your application adds them to a model's prompt to produce an answer.

HydraDB also builds a **context graph**, a record of relationships such as "the Payments team owns the billing service." Search results can include those relationships when a question needs connected information. See [Context Graphs](/essentials/v2/context-graphs) when you need this detail.

***

## Performance

* **Memory recall:** 90.79% on LongMemEval-S, a benchmark for remembering information from past conversations.

View the full technical breakdown in our [benchmarks](https://benchmarks.hydradb.com/).

***

## What you can build

* Customer support agents grounded in real customer history and preferences
* Coding agents with persistent memory of a codebase and team conventions
* Clinical companions tracking patient context across visits
* Research copilots reasoning across papers, authors, and findings
* Internal knowledge assistants spanning Slack, Notion, Drive, and email
* Personal assistants that remember preferences across conversations

Find more use cases at [Cookbooks](/cookbooks/v2/index).

***

## Get started

1. Sign up at [app.hydradb.com](https://app.hydradb.com) for your API key
2. Follow the [Quickstart](/get-started/v2/quickstart) to save and retrieve your first memory
3. Explore [Core Concepts](/get-started/v2/core-concepts) and the [API Reference](/api-reference/v2)

For enterprise onboarding, contact [founders@hydradb.com](mailto:founders@hydradb.com).

## For AI agents

For AI coding agents and IDE assistants, use the [HydraDB Agent Integration Guide](/AGENTS) and the [v2 OpenAPI spec](/api-reference/v2/openapi.json).


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.