FAQ
Short answers about the platform, the course model, and how Knowledge.Dev fits into an AI-assisted engineering workflow.
What is Knowledge.Dev?
Knowledge.Dev is an AI-native learning platform for developers who want to build real systems while working with an AI coding agent.
The platform is built around structured playgrounds, local tooling, and an agent-guided learning flow.
How do I start?
The fastest preview is Demo Prompt.
The full learning flow starts with Quick Start, where you install the launcher, connect the MCP Server, and open a playground in your agent.
Do I need an AI agent?
Yes.
Knowledge.Dev is designed around an AI coding agent that can work in a local folder, read the learning context, and guide the session step by step through the Playground Workflow.
What is a playground?
A playground is a complete guided engineering course built around a real project.
Each playground has its own scope, structure, and unique ID. You choose one from the catalog, open it in your agent, and build the project locally through the Playground Workflow.
How is a playground different from the demo prompt?
The demo prompt is a lightweight preview.
The Playground Workflow is the full course experience, with MCP-backed lesson context, more reliable progress flow, and access to the complete course structure.
What does the launcher do?
The Launcher prepares Knowledge.Dev on your machine.
It installs the local tooling, downloads the correct binary build for your system, and acts as the entrypoint for MCP-related commands.
What does the MCP Server do?
The MCP Server connects Knowledge.Dev to your AI agent.
It provides the lesson context, structure, and course flow that allow the agent to guide the session in a controlled way.
What is the difference between the launcher and the MCP Server?
The Launcher is the local command you install and run.
The MCP Server is the part that the agent connects to during the learning session. In practice, launcher commands are used to prepare and work with that MCP layer.
What is the learning approach?
Knowledge.Dev teaches through guided construction.
Instead of passively reading or watching, you work inside a real project, follow the AI tutor, write code locally, and assemble the system step by step.
Are these toy examples or real projects?
The playgrounds are structured as real applications and engineering exercises, not isolated fragments.
The goal is to help you build working systems that feel like actual development, not just rehearse tiny disconnected snippets.
Do I need a subscription?
You can explore public parts of the platform without going through the full paid workflow.
For full course access, the MCP-backed experience, and the broader library tied to your account, you should expect to sign in with an account that has the required access.
Is everything available today?
No.
Some parts of the platform are still evolving, especially the Workbench tools and some surrounding product surfaces.
Where should I go after this page?
If you want the full setup, continue with Quick Start.
If you only want a fast first look, open Demo Prompt.
If you have questions, visit our Discord