AI Agents from Zero: Practical Bootcamp to Enterprise Deployment
A systematic, open-source Chinese tutorial covering tutorials, runnable source code, real-world projects, and interview prep for AI agent development.
- Source repo
- didilili/ai-agents-from-zero
- Stars
- ★ 5.4k
- Last updated
- 5d ago
- License
- MIT
- Primary language
- Python
- FA score
- 34/100 · Major gaps
At a glance
- How it runs
- Works with
- Portable with changesOpenAI API · Claude API
- Cost
- Free software; you pay for model usage
- Setup effort
- Medium · a few setup steps
- You'll need
- Typical use
- Self-learners who want a systematic path into AI agent development without paying for expensive bootcamps.
- Not a fit if
- Readers wanting a quick skim rather than a long systematic course
- Developers who prefer the Java stack (Spring AI / langchain4j)
- Non-Chinese readers needing a fully English tutorial
- Source review
- 34/100 · Major gaps 8 safety controls not found
What does this agent do, and when should you use it?
ai-agents-from-zero is a long-maintained open-source Chinese tutorial aimed at taking learners from zero to enterprise-level proficiency in AI agents and LLM application development. It covers fundamental concepts, frameworks like LangChain and LangGraph, low-code platforms such as Coze and Dify, protocols including MCP and A2A, plus enterprise-grade RAG/agent projects and an interview question bank. The tutorial emphasizes "run-able" examples, providing runnable source code and environment guides for each case. It includes two complete real-world projects: an NL2SQL + LangGraph project (ecommerce question answering) and a DeepAgents multi-agent project (deep research and search), with source code in separate repositories. Licensed under MIT, the project is continuously updated and restructured from public course materials. Follow-along learners need Python 3.10+ and API keys, with options for local Ollama models.
This repository is a comprehensive tutorial collection, not a standalone agent application. It provides structured chapter notes, runnable case code, project documentation, and an interview question bank. Learners clone the repo, set up a Python 3.10 environment, configure API keys (e.g., Qwen, DeepSeek), and run example scripts from the root, such as python 案例与源码-2-LangChain框架/01-helloworld/StandardDesc.py. The ecommerce project (shopkeeper-agent) integrates MySQL, Qdrant, Elasticsearch, LangGraph, and FastAPI for natural-language-to-SQL generation, while the deep research project (deepsearch-agents) uses DeepAgents for multi-agent orchestration, web search, knowledge bases, and file generation. The tutorial also covers Docker deployment, fine-tuning with Llama-Factory, and MCP/A2A protocols.
- Self-learners who want a systematic path into AI agent development without paying for expensive bootcamps.
- Developers from frontend, backend, or product backgrounds aiming to transition to LLM application engineering.
- Engineers needing to implement enterprise-grade RAG solutions, referencing cases like the ecommerce Q&A agent.
- Job seekers preparing for AI developer interviews, using the curated question bank aligned with job descriptions.
- Technical staff exploring low-code platforms and protocols like Coze, Dify, MCP, and A2A.
- Users who prefer local model deployment, following guides for Ollama, Xinference, and avoiding cloud API reliance.
How do you install or deploy this agent?
Clone the repo: git clone https://github.com/didilili/ai-agents-from-zero.git && cd ai-agents-from-zero. Create a virtual environment (Python 3.10 recommended): python3.10 -m venv .venv && source .venv/bin/activate (Windows: .venv\Scripts\activate), then pip install -r requirements.txt. Copy .env-example to .env and add your API keys (e.g., QWEN_API_KEY), or use Ollama for local models without keys.
How do you use this agent?
Run the first example from the project root: python 案例与源码-2-LangChain框架/01-helloworld/StandardDesc.py. Ensure you are in the root directory to load .env. For more examples and real-world projects, refer to the online documentation and the tutorial outline.
What are this agent's strengths and limitations?
- Extremely systematic, covering the complete path from fundamentals to enterprise deployment, with continuous updates.
- Every case provides runnable source code and environment guidance, ensuring real-world usability and reducing pitfalls.
- Includes two complete production-grade projects (ecommerce Q&A and deep research), with reusable source code ideal for resumes.
- Interview question bank organized by skill domains, compiled from real big-tech interview questions.
- Focused on Python + LangChain/LangGraph, aligning with mainstream engineering paths and avoiding Java-stack bias.
- The tutorial is massive and requires significant time investment, not suited for quick-start seekers.
- Relies on multiple external services and platforms (API keys, vector DBs, cloud deployments), adding configuration overhead.
- Project source code is in separate repositories, requiring extra fetching and potentially specific data environments.
- Some content is based on paid courses, though credited and restructured; originality may be limited.
- Primarily for Chinese-speaking readers, with limited English support, which may hinder non-Chinese users.
How does this agent compare with similar options?
The tutorial contrasts with fragmented online posts, paid bootcamps, and Java-based stacks like Spring AI and langchain4j, emphasizing systematicity, open-source, and Python as key differentiators.
Key facts side by side with the most closely related agents.
| Agent | Source review | Form / cost | Stars | Updated | Language | Full support on |
|---|---|---|---|---|---|---|
| AI Agents from Zero: Practical Bootcamp to Enterprise Deployment This agent | 34 · Major gaps | CLIFree + model costs | ★ 5.4k | 5d ago | Python | OpenAI API · Claude API |
| Dive into LangGraph | 69 · Some gaps | Library / SDKFree + model costs | ★ 461 | 1mo ago | Jupyter Notebook | Claude Code |
| AI Agent Engineering Textbook | 55 · Major gaps | FrameworkFree | ★ 601 | 2d ago | HTML | — |
| AgentGuide - All-in-One AI Agent Development & Interview Guide | 23 · Major gaps | Library / SDKFree + model costs | ★ 11k | 1mo ago | MDX | OpenAI API · Claude API |
How does FollowAgents rate this agent?
Why each dimension lost points
Evidence shows: repository is tutorial-oriented, no malicious code or permission requests found, but no security mechanism documentation. No least privilege, user confirmation, data flow transparency, sensitive data handling, dependency security, external effects, rollback, or source attribution implementations or documentation. Therefore all trust criteria score 0.
Self-consistency: README, outline, case links, and changelog are consistent, tutorial structure clear, score 2. Dependency availability: requirements.txt lists clear dependencies and version ranges, but actual availability not verified, score 2. Failure messages: newbie guide mentions common error handling, but no specific error message examples, score 1.
Audience and scenarios: clearly targets beginners to engineers, covers multiple scenarios, score 2. Capability boundaries: tutorial scope clear, but no mention of what is not covered, score 1. Trigger precision: not applicable, score 0. Environment fit: provides Python version, virtual environment, API key configuration, etc., score 2.
Information architecture: clear table of contents and navigation, score 2. Install notes: detailed installation steps, score 2. Naming stability: no naming conventions mentioned, score 1. Examples and FAQ: provides cases and common questions, score 2. Known limitations: not explicitly listed, score 1. License: MIT license, score 2. Versioning and changelog: has changelog but no version numbers, score 1. Maintenance responsibility: no explicit maintainer, score 1.
Output usability: tutorial content well organized, but no directly runnable output examples, score 1. Marginal value: provides systematic tutorial and interview question bank, high value, score 2. Cost-benefit: free and open source, but requires self-configuration of API keys, score 1.
Claim traceability: some claims mention reference courses and interview question sources, but no specific links, score 1. Cross-source corroboration: no external verification, score 1. Fact-inference separation: not explicitly distinguished, score 1.
- Not found in source: least-privilege scopingGrant only what the task needs: a dedicated account or read-only token, scoped to specific directories and repos.
- Not found in source: confirmation before actingTurn on (or add) a confirmation step before it acts, and try it in a sandbox or test environment before real data.
- Not found in source: data-flow disclosureWatch which external services it contacts (proxy or firewall logs) and keep sensitive data out until you know where it goes.
- Not found in source: sensitive-data handlingUse dedicated, low-privilege, revocable API keys — never production credentials — and keep secrets out of logs.
- Not found in source: dependency securityPin versions and run a dependency audit (npm audit, pip-audit) before installing; prefer running it in a container.
- Not found in source: disclosed external effectsEstablish which external systems it writes to, sends to or changes, and verify with test accounts or repos before production.
- Not found in source: rollback or recovery pathBack up first, or work on a git branch or snapshot, so its changes can be undone.
- Not found in source: verifiable attributionInstall from the official repo or registry and check the publisher and URL to avoid look-alike packages.
- Repository is tutorial-oriented, no executable code or security mechanisms provided, users need to assess risks themselves.
- Many dependencies, actual availability not verified, may encounter compatibility issues during installation.
- No explicit version numbers, changelog may be incomplete.