Skip to main content
Conversational Agentic Platform

AI agents that chat for your customers

Configure intelligent agents, connect them to WhatsApp and more channels. With RAG over your documents and external tools. All on a secure multi-tenant platform.

Start for free See how it works
Features

Everything you need to operate AI agents

A complete platform to configure, connect and manage intelligent conversational agents.

Intelligent AI Agents

Configure LLM models (OpenAI, Claude), adjust temperature, customize system prompts. Each agent adapts to your use case.

Multichannel

Connect with WhatsApp, Telegram, Web, SMS, Email and Voice from a single place. Your customers pick the channel, you manage everything centrally.

RAG / Knowledge Base

Ingest PDF, DOCX, CSV and more. Vector search over your documents so the agent answers with accurate information from your organization.

Tools and Integrations

Connect REST APIs, gRPC services and n8n workflows. The agent triggers real actions in your systems.

Secure Multi-Tenant

Complete isolation per organization with Row Level Security in PostgreSQL and RBAC with Casbin. Each tenant operates in its own secure space.

Dashboard and Analytics

Metrics and KPIs tailored by role. Super admins see everything, admins their tenant, agents their conversations.

Process

How Aruka works

Four steps to have your conversational agent up and running.

1

Configure

Define your agent: LLM model, prompts, temperature and custom behavior.

2

Connect

Wire up communication channels (WhatsApp, Web, etc.) and external tools.

3

Converse

Your agent handles conversations in real time with context from your documents.

4

Analyze

Monitor performance, satisfaction and key metrics from the dashboard.

Integrations

Integrates with your tools

Connect Aruka with the services you already use. Communication channels, AI models and external tools.

WhatsApp
Telegram
OpenAI
Claude
PostgreSQL
Firebase
n8n
REST API
gRPC
Web Chat
Slack
Email SMTP
Pricing

Plans that adapt to your business

Pick the plan that fits your needs. Scale up when you need to.

Starter

For teams starting with conversational agents.

$49 /mo
  • Up to 1,000 conversations/month
  • 1 communication channel
  • 1 configured agent
  • 100 MB of RAG documents
  • 5 external tools
  • Email support
Get started
Popular

Pro

For companies that need to scale their operations.

$149 /mo
  • Up to 10,000 conversations/month
  • 5 communication channels
  • 5 configured agents
  • 1 GB of RAG documents
  • 25 external tools
  • Advanced dashboard
  • Priority support
Get started

Enterprise

For organizations with bespoke needs.

Custom
  • Unlimited conversations
  • Unlimited channels
  • Unlimited agents
  • Custom storage
  • Unlimited tools
  • SSO and SAML
  • Dedicated SLA
  • 24/7 support
Contact sales
FAQ

Frequently asked questions

Answers to the most common questions about Aruka.

What is Aruka and who is it for?
Aruka is a multi-tenant conversational agentic platform designed for organizations that want to operate AI agents. It lets you configure intelligent agents, connect them to multiple communication channels, integrate your own documents via RAG, and connect external tools. Ideal for B2B companies that need to automate customer support with AI.
Which communication channels does it support?
Aruka currently supports WhatsApp as the primary channel with HMAC-SHA256 signature validation. You can also configure Telegram, Web Chat, SMS, Email and Voice. Each channel is managed independently per tenant, and you can enable or disable channels as needed.
How does RAG over my documents work?
Aruka ingests documents in multiple formats (PDF, DOCX, CSV, XLSX, HTML, Markdown) up to 50 MB. Documents are processed automatically: split into chunks, vector embeddings generated with OpenAI, stored in PostgreSQL with pgvector. When a user asks a question, the agent automatically searches your knowledge base by semantic similarity and answers with accurate information.
Is it secure for multiple organizations?
Yes. Aruka implements multi-tenant isolation at every layer: Row Level Security (RLS) in PostgreSQL for transactional data, RBAC with Casbin for access control, path-based isolation in Firestore for configuration, Fernet-encrypted credentials per tenant, and per-organization rate limiting at the API Gateway. Each tenant operates in a fully isolated space.
Can I connect my own APIs and tools?
Yes. Aruka supports three connector types for external tools: REST APIs (with GET, POST, PUT, DELETE methods), gRPC services (with host, port, service and method configuration), and n8n workflows (via webhook URL). Each tool is defined with a JSON Schema input for automatic validation, and the agent decides when to call them during the conversation.
Which AI models can I use?
Aruka lets you choose between OpenAI models (GPT-4o, GPT-4o-mini, etc.) and Anthropic Claude (Opus, Sonnet, Haiku). You can configure the model, temperature (creativity), max tokens and system prompt for each agent independently per tenant.