Documentation

Self-Hosted Deployment on Your Infrastructure

For corporations with strict PII and regulatory requirements, Tridan AI can be deployed entirely inside your VPC or on-premises data centre. You get the complete platform with unlimited AI agents, full customisation, and total data sovereignty — priced per project as a white-glove engagement.

Enterprise-Grade Privacy & Compliance

Designed for organisations where PII protection and data-protection rules such as the GDPR and CCPA are mandatory. Your data never leaves your infrastructure, which makes regulatory compliance dramatically simpler.

Why Self-Host?

Total Data Sovereignty

Your sensitive data never leaves your infrastructure. No external servers, no third-party cloud dependencies — only you have access to your information.

Compliance by Design

Built for organisations where PII protection and data-protection rules such as the GDPR and CCPA are non-negotiable.

Unlimited Scale

No limits on users, requests, or AI agents. Support your entire organisation without usage restrictions or per-seat costs.

Full Customisation

Integrate with your existing systems, customise workflows, and white-label the platform to match your brand.

The Full Stack, Inside Your Network

Every component of the platform runs in your environment — there is no callback to Tridan AI servers.

Qdrant Vector Database

Semantic search and retrieval running entirely inside your environment.

Local Ollama Models

Run open models locally so prompts and answers never leave your network, or connect your own API keys for OpenAI, Anthropic, and other providers.

Self-Hosted Supabase

Authentication, storage, and application data on your own database instance.

PII Detection

Built-in detection of personally identifiable information to protect sensitive data in every conversation.

What's Included

  • Full platform deployment on your VPC or on-premises servers
  • Your own LLM API keys, or fully local models via Ollama
  • Connections to static and dynamic data sources
  • Unlimited role-specific AI agents for every team and department
  • Complete white-labelling: your logo, colours, and domain
  • Full DevOps support and monitoring for the contract duration
  • Custom feature development by a dedicated engineering team
  • Training and documentation handover for your staff

Implementation Process

  1. 1

    Consultation & Planning

    We assess your infrastructure, compliance needs, and integration goals to create a customised deployment plan.

  2. 2

    Infrastructure Setup

    Our team works with your IT department to deploy the full stack on your infrastructure with optimal configuration and security.

  3. 3

    Integration & Configuration

    Connect your model keys or local models, integrate existing data sources, and configure role-specific AI agents.

  4. 4

    Training & Go-Live

    Comprehensive training for your team, documentation handover, and ongoing support for successful adoption.

Technical Requirements

Minimum Infrastructure

  • CPU: 8+ cores (16+ recommended)
  • RAM: 32 GB minimum (64 GB+ recommended)
  • Storage: 500 GB SSD (1 TB+ recommended)
  • Network: Stable internet only if using external LLM APIs
  • OS: Linux (Ubuntu 20.04+ or CentOS 8+)

Security Requirements

  • Firewall: Configurable for restricted access
  • SSL/TLS: HTTPS encryption required
  • Database: Encrypted storage support
  • Backup: Regular automated backups
  • Monitoring: System health monitoring tools

Prefer a managed subscription instead? Compare the cloud plans on All Plans Overview.

Ready to Deploy?

Talk to our enterprise team about your infrastructure, compliance requirements, and timeline for a custom deployment.

Book an Enterprise Consultation

Last updated · Reviewed by Nemanja Milivojevic, Founder