# Use FastGPT

> Master FastGPT core concepts, hands-on tutorials, deep guides, and practical cases to build enterprise-grade AI applications.

Source: https://doc.fastgpt.io/en/guide/getting-started

FastGPT is an AI application orchestration platform providing out-of-the-box model orchestration, accurate knowledge base retrieval, autonomous Agent planning, and visual workflow development.

## Quick Start and Tutorials

Follow structured tutorials to build operational AI systems in 5 to 10 minutes:

* [5-Minute Quick Start](/en/guide/getting-started/quick-start): Import documents, connect a Chat Agent, test responses, and generate a public share link.
* [Build Chat Agent V2](/en/guide/getting-started/first-agent): Configure an autonomous agent with dynamic planning, tool calling, and multi-step reasoning.
* [Build Your First Workflow](/en/guide/getting-started/first-workflow): Master canvas connections, variable references, response output modes, and end-to-end testing.

## Core Concepts and Architecture (Explanation)

Understand system design, retrieval mechanics, and selection criteria (aligned with the "Core Concepts" sidebar section):

* [Core Concepts and Topology](/en/guide/getting-started/concepts): Understand applications, collections, chunks, tools, Skills, version snapshots, and permissions.
* [Application Types and Selection](/en/guide/getting-started/application-types): Compare Chat Agent, Chat Agent V2, and Workflow across execution models and decision trade-offs.
* [Workflow Orchestration Overview](/en/guide/build/workflow/intro): Understand node scheduling, edge state transitions, loops, parallel execution, and resumption.
* [RAG Principles and Retrieval Architecture](/en/guide/dataset/rag): Master vector recall, full-text search, RRF hybrid fusion, and Rerank scoring.

## Practical Guides and Case Studies (How-to Guides)

Follow standardized recipes and enterprise blueprints for concrete business goals:

* [Accurate Knowledge-Base Q\&A](/en/guide/getting-started/knowledge-qa): Master text chunking, vector retrieval with reranking, prompt constraints, and citations.
* [Lesson 1: Corporate Email Assistant](/en/guide/getting-started/examples/email-assistant): Build a structured business email generator with Chat Agent instructions and tool extensions.
* [Lesson 2: Civil Code Q\&A Assistant](/en/guide/getting-started/examples/civil-code-qa): Combine document chunking, vector retrieval, and grounded prompts for legal clause lookup.
* [Lesson 3: Content Review Workflow](/en/guide/getting-started/examples/content-review): Orchestrate question classification, AI rewriting, and human confirmation branches on the canvas.
* [Lesson 4: Data Analysis Agent V2](/en/guide/getting-started/examples/data-analysis): Run Python aggregation and chart generation inside the virtual machine sandbox.

## Role-Specific Entry Points

* [Developer Portal](/en/developers): Integrate through the API or explore [plugin development](/en/plugin/intro) for tools, model presets, debugging, and publishing.
* [Deployment and Operations](/en/self-host): Deploy via Docker or Sealos, configure vector stores and models, monitor workloads, and upgrade versions.
* [Knowledge Base Lifecycle](/en/guide/dataset/lifecycle): Study text chunking, indexing maintenance, incremental sync, and tag management.
* [Extensions and Tool Ecosystem](/en/guide/build/tools/overview): Connect HTTP tools, system plugins, code execution sandboxes, and MCP services.

## Community and Technical Support

Documentation follows the current user interface. In self-hosted environments, capability availability depends on version licensing, infrastructure services, team resource permissions, and backend model support.

Report technical issues through [GitHub Issues](https://github.com/labring/FastGPT/issues/new/choose). When reporting bugs, provide the software version, step-by-step reproduction instructions, expected outcomes, and sanitized log output.