Skip to main content

​
Quick start

This guide will help you deploy your first voice agent with Iqra AI. You can choose between Iqra Cloud (fully managed) or self-hosted deployment.
New to Iqra AI? Start with Iqra Cloud to get up and running in minutes without infrastructure setup.
Iqra Cloud is the fully managed, production-ready platform with multi-tenant billing, whitelabeling management, and automatic scaling.
1

Create an account

Sign up for a free account at app.iqra.botYou’ll get immediate access to:
  • Visual IDE for building conversation flows
  • Pre-configured AI models and voice providers
  • Global edge network with multi-region routing
  • Production-ready infrastructure
2

Create your first agent

From the dashboard:
  1. Click Create New Agent
  2. Choose a template or start from scratch
  3. Configure your agent’s base settings:
    • Name: Give your agent a descriptive name
    • Language: Select primary language(s)
    • Voice provider: Choose from ElevenLabs, Deepgram, Azure Speech
    • LLM provider: Select OpenAI, Anthropic, Gemini, or Groq
3

Build your conversation flow

Use the Visual IDE to design your agent’s behavior:
  1. Open the Script Builder
  2. Add conversation nodes by dragging from the left panel
  3. Configure system prompts and AI instructions
  4. Add deterministic workflows for business logic:
    • Use If/Else nodes for conditional routing
    • Add Loops for repetitive tasks
    • Insert Variables to maintain conversation state
  5. Connect nodes to define the conversation flow
The Visual IDE exposes granular control while remaining accessible to non-engineers. See Building Scripts for detailed guidance.
4

Add integrations (optional)

Connect your agent to external services:
  1. Navigate to Tools in the left sidebar
  2. Browse available FlowApps (Cal.com, HubSpot, etc.)
  3. Click Configure and add your API credentials
  4. Use the integration in your script by adding a FlowApp Action node
FlowApps abstract external APIs so you can configure them visually without writing HTTP requests.
5

Test your agent

Test your agent before deployment:
  1. Click Test in the top-right corner
  2. Choose your test channel:
    • Browser call: Test via WebRTC directly in your browser
    • Phone number: Get a temporary test number
  3. Have a conversation with your agent
  4. Review the Debug Console to see:
    • Conversation transcripts
    • Variable states
    • Execution flow
    • API call logs
6

Deploy to production

When you’re ready to go live:
  1. Click Deploy
  2. Choose your deployment channel:
    • SIP trunking: Connect via Twilio, Telnyx, or Vonage
    • WebRTC/WebSocket: Embed in your website or mobile app
    • Direct phone number: Get a dedicated phone number
  3. Configure your production settings
  4. Click Activate
Your agent is now live and handling real conversations!

​
Option 2: Self-hosted deployment

Run the core engine on your own infrastructure. This version includes the full Agent Engine, Script Builder, and FlowApp system, but excludes the commercial billing and whitelabeling modules.
This codebase is currently in pre-release (v0.1 pending). Automated database seeding scripts are not yet available. Production deployment requires manual database setup.

​
Prerequisites

Before deploying, ensure you have:
  • .NET 10 Runtime installed
  • MongoDB for data persistence
  • Redis for caching and session management
  • Milvus for vector storage (embeddings)
  • RustFS or S3-compatible storage for media files

​
Installation steps

1

Clone the repository

git clone https://github.com/abdofallah/IqraAI.git
cd IqraAI
2

Configure environment

Create your appsettings.json with required services:
{
  "ConnectionStrings": {
    "MongoDB": "mongodb://localhost:27017/iqraai",
    "Redis": "localhost:6379"
  },
  "Milvus": {
    "Host": "localhost",
    "Port": 19530
  },
  "Storage": {
    "Provider": "S3",
    "Endpoint": "your-s3-endpoint",
    "Bucket": "iqraai-media"
  }
}
Never commit API keys or secrets. Use environment variables for sensitive configuration.
3

Set up database (manual)

Automated seeding scripts are pending v0.1 release. For now, you’ll need to manually configure the database schema.
Refer to the Self-Hosting Guide for detailed database setup instructions.
4

Build and run

dotnet restore
dotnet build
dotnet run --project IqraBackendApp
The backend will start on https://localhost:5001 by default.
5

Access the dashboard

Open your browser and navigate to https://localhost:5001You can now build agents using the Visual IDE, just like with Iqra Cloud.

​
Next steps for self-hosted

Self-hosting guide

Complete deployment guide with production configurations

System architecture

Understand the component separation and dependencies

​
What’s next?

Now that you have your first agent running:

Build advanced scripts

Learn how to use deterministic workflows and conditional logic

Multi-language support

Configure parallel context stacks for native multilingual agents

Secure sessions

Implement PCI-DSS compliant data collection

Smart interruptions

Configure turn-taking and barge-in detection

​
Get help

Build docs developers (and LLMs) love

Get started for freeTalk to us