Documentation

Everything you need to know

Guides, architecture references, and FAQs for the Chervik platform.

Vision

Why we built Chervik and the problem we're solving for enterprise AI.

Architecture

Technical overview of the Chervik platform architecture and design decisions.

Product Concepts

Core concepts: knowledge ingestion, RAG, hybrid search, agents, and workflows.

Getting Started

How to deploy Chervik in your environment and configure your first knowledge base.

FAQs

Answers to common questions about deployment, security, and integrations.

FAQ

Frequently Asked Questions

Does Chervik send data to external AI providers?+

No. Chervik is designed to run entirely within your infrastructure. You can use local models (Llama, Mistral) or private API endpoints. No data is sent to external services without your explicit configuration.

What cloud providers does Chervik support?+

Chervik runs on any Kubernetes-compatible environment — AWS, Azure, GCP, or on-premise. Infrastructure is defined as Terraform modules for easy deployment.

How long does deployment take?+

A standard deployment takes 3–7 days for a pilot environment. Production deployments with full security hardening typically take 2–4 weeks.

What document formats are supported?+

PDF, Word, Excel, PowerPoint, Markdown, HTML, plain text, and any format with a custom parser. Confluence, SharePoint, and Notion connectors are available.

Is Chervik GDPR compliant?+

Chervik's architecture supports GDPR compliance — data stays in your infrastructure, access is controlled, and deletion requests can be processed. Compliance depends on your specific configuration and policies.

Can Chervik integrate with our existing SSO?+

Yes. Chervik supports SAML 2.0 and OIDC for SSO integration with Okta, Azure AD, Google Workspace, and any standards-compliant identity provider.