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

# Overview

> Overview of Sayvy AI's voice engine, WhatsApp automation, in-built CRM, and AI memory

# Overview

Sayvy AI unifies Web Voice AI agents, WhatsApp Business automation, a native in-built CRM, and long-term conversational memory into a cohesive real-time conversational intelligence platform.

***

## Architecture

```mermaid theme={null}
graph TD
    subgraph Channels ["Channels"]
        Web[Web Audio / WebRTC]
        WA[WhatsApp Cloud API]
    end

    subgraph Core ["Engine"]
        Voice[Voice AI]
        WhatsApp[WhatsApp Engine]
    end

    subgraph Data ["Data Layer"]
        CRM[Smart CRM]
        Memory[AI Memory]
    end

    Web --> Voice
    WA --> WhatsApp
    WA -.-> Voice
    Voice <--> CRM
    WhatsApp <--> CRM
    Voice <--> Memory
    WhatsApp <--> Memory
    CRM <--> Memory
```

***

## The 4 Pillars

<CardGroup cols={2}>
  <Card title="Voice AI" icon="microphone" href="/documentation/voice/web-audio-webrtc">
    Browser WebRTC voice calls, WhatsApp audio, interruptions, and voice personas.
  </Card>

  <Card title="WhatsApp" icon="whatsapp" href="/documentation/whatsapp/setup">
    Meta Cloud API setup, messaging window, templates, and interactive buttons.
  </Card>

  <Card title="Smart CRM" icon="address-book" href="/documentation/crm/data-model-and-pipelines">
    Native contacts, deal pipelines, custom attributes, and interaction timelines.
  </Card>

  <Card title="AI Memory" icon="brain" href="/documentation/memory/identity-resolution">
    Cross-channel identity resolution, context buffers, semantic memory, and knowledge base.
  </Card>
</CardGroup>


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