Documentation

Introduction

KndTech AI is a context-aware chatbot that runs on your company's documents and knowledge base. You connect sources (Confluence, Google Drive, PDFs, URLs), group them into collections, and chat with answers that cite the exact snippets used. Each workspace can have its own collections and members, so support, engineering, sales, and HR each get an assistant tuned to their content.

The docs below cover getting started, managing workspaces and knowledge, how chat and citations work, admin and security, and usage and billing. If you're new, start with Getting started.

Getting started

You can go from signup to a working assistant in under an hour. Here's the typical path.

1. Create an account. Sign up at the app; you get a personal workspace by default.
2. Create a collection. In the app, add a collection and give it a name (e.g. "Support FAQs" or "Engineering runbooks").
3. Add sources. Upload PDFs, paste URLs, or connect Confluence or Google Drive if your plan supports it. We index the content and chunk it for retrieval.
4. Chat. Open a chat in that workspace and ask questions. Answers are grounded in the collection's documents and include citations you can click to open the source.

For teams, invite members from the workspace settings and assign roles so some people can manage collections and others can only chat. See Workspaces & teams for details.

Workspaces & teams

A workspace is the container for members, collections, and chats. You can have one workspace per team or use a single workspace with multiple collections. Members are invited by email; roles control who can add or edit sources, create collections, or only use chat.

Workspace settings let you rename the workspace, manage members and roles, and (on paid plans) configure SSO. All usage and billing for that workspace are visible to admins in the dashboard.

Knowledge & collections

Content is organized into collections. Each collection has a set of sources (documents, URLs, or connected integrations). When you chat, the model retrieves relevant chunks from the collection's index and uses them to generate an answer. That keeps responses accurate and cited.

• Supported sources: PDF uploads, URLs, and (on supported plans) Confluence spaces and Google Drive. We chunk and embed text; images in PDFs are not indexed.
• Chunking: Documents are split into overlapping segments so retrieval can pull the right context. You don't configure chunk size; we use defaults tuned for RAG.
• Updates: When you add or update a source, we re-index so new content appears in answers. Deleted sources are removed from the index.

Chat & citations

Chat uses the collection(s) attached to the conversation. You ask in natural language; the system finds relevant chunks, sends them to the model, and the model replies with an answer and inline citations. Clicking a citation opens the source document or snippet.

For best results, ask clear, specific questions. Broad or vague prompts may still return an answer, but citations are most useful when the question matches the kind of content in the collection (e.g. "What is our refund policy?" for a support collection, or "How do we deploy the API?" for runbooks).

Admin & security

Admins can manage workspaces, members, roles, and (on Enterprise) SSO and audit logs. Data is encrypted in transit and at rest; we don't train on your content. Access is role-based so only authorized users see the right collections and chats.

For security and legal review, see our Privacy Policy and Terms of Service. Enterprise customers get additional controls and compliance documentation; contact sales for details.

Usage & billing

Plans define seat limits, number of collections, and message caps per month. Usage is visible in the dashboard: messages sent, tokens consumed, and (where applicable) spend. Staying within plan limits avoids overages; upgrading or contacting sales unlocks higher limits and Enterprise features.

For exact limits and pricing, see Pricing. Questions about billing or custom plans can go to in-app support or Contact.

Support & resources

• In-app help: Use the help menu inside the app for the fastest response. Include your workspace name and a short description of the issue.
• Contact form: For demos, enterprise inquiries, or partnership, use Contact. We reply within 24 hours on business days.
• Privacy & terms: Privacy Policy, Terms of Service.

Don't have an account yet? Sign up to get started.