# What is ParticleDB?

A developer introduction to relational data, key-value, vectors, analytics and Particles in ParticleDB.

Source: https://particledb.ai/getting-started/introduction/

ParticleDB is a database for AI agents and applications. It combines SQL, key-value access, vector search, analytics and functions called Particles in one database engine.

Use it to store application records, retrieve context, track work and query operational data. Your application controls the context an agent receives and the actions it may take.

## The building blocks

| Need | Start here |
|---|---|
| Records, relationships and transactions | [Relational data](/guides/relational-data/) |
| Keys, expiration and coordination | [Key-value](/guides/key-value/) |
| Relevant context and embeddings | [Search and memory](/guides/search-and-memory/) |
| Reports on application records | [Analytics](/guides/analytics/) |
| Custom logic beside the data | [Particles](/particles/overview/) |
| Agent state and application workflows | [Agents and workflows](/guides/agents-and-workflows/) |

## Choose a starting point

New installation: follow the [quickstart](/getting-started/quickstart/).

Existing application: choose a [language client](/sdks/overview/) or read the [migration guide](/operations/migration/).

Looking up behavior: use the [SQL reference](/sql/overview/) and [limits](/limits/overview/).

## Data and execution boundaries

Data in other services needs an explicit import or integration. Storing data in one engine does not automatically authorize an agent to access it. Application workers and external tools also need their own retry and idempotency handling. [Security](/guides/security/) and [workflow patterns](/guides/agents-and-workflows/) explain these boundaries.

## Related

[Documentation home](/docs/) · [Troubleshooting](/guides/troubleshooting/)