What is this?
This is a comprehensive, long-term-maintained AI Agent / LLM application development tutorial in Chinese, covering everything from basic concepts to enterprise-grade deployment, with runnable projects and an interview question bank.
This is a comprehensive, long-term-maintained AI Agent / LLM application development tutorial in Chinese, covering everything from basic concepts to enterprise-grade deployment, with runnable projects and an interview question bank.Is this a coding framework or a learning curriculum?It is a structured curriculum/course repository, not a software library: it walks through concepts, low-code platforms (Coze, Dify), code frameworks (LangChain, LangGraph), RAG/Agent projects and fine-tuning, with runnable example code for each topic.
It is a structured curriculum/course repository, not a software library: it walks through concepts, low-code platforms (Coze, Dify), code frameworks (LangChain, LangGraph), RAG/Agent projects and fine-tuning, with runnable example code for each topic.How do I run the example projects?Clone the repo, set up a Python 3.10–3.13 environment, then copy .env-example to .env, fill in your API key, and run the example script from the project root directory.
Clone the repo, set up a Python 3.10–3.13 environment, then copy .env-example to .env, fill in your API key, and run the example script from the project root directory.Do I need an API key from a model provider to try the examples?Yes, you need an API key for a model like Qwen (通义千问/阿里百炼) or DeepSeek set in the .env file, unless you run a local model through Ollama, which needs no key.
Yes, you need an API key for a model like Qwen (通义千问/阿里百炼) or DeepSeek set in the .env file, unless you run a local model through Ollama, which needs no key.Any gotchas when running the scripts?Python scripts must be executed from the project's root directory, or they won't find the .env file and API key configuration, causing ModuleNotFoundError or auth errors.
Python scripts must be executed from the project's root directory, or they won't find the .env file and API key configuration, causing ModuleNotFoundError or auth errors.Who is this curriculum aimed at?It targets people aiming to become AI application/Agent development engineers, especially those switching from front-end, back-end, or product roles into AI, as well as job-seekers preparing for related interviews.
It targets people aiming to become AI application/Agent development engineers, especially those switching from front-end, back-end, or product roles into AI, as well as job-seekers preparing for related interviews.What real-world projects are included?Completed hands-on projects include an e-commerce Q&A system (NL2SQL + LangGraph, with MySQL, Qdrant and Elasticsearch) and a multi-agent deep research system built on DeepAgents with web search and file generation.
Completed hands-on projects include an e-commerce Q&A system (NL2SQL + LangGraph, with MySQL, Qdrant and Elasticsearch) and a multi-agent deep research system built on DeepAgents with web search and file generation.Why Python instead of Java frameworks like Spring AI?The author deliberately focuses on the Python + LangChain + LangGraph stack because it is the most common engineering path for AI Agent and LLM application development, unlike courses that lean on Spring AI or langchain4j.
The author deliberately focuses on the Python + LangChain + LangGraph stack because it is the most common engineering path for AI Agent and LLM application development, unlike courses that lean on Spring AI or langchain4j.What does the interview question bank cover?It is organized by job competency areas matching real JD requirements, with questions compiled from actual big-tech interviews and common follow-up scenarios, useful for job switchers and new graduates.
It is organized by job competency areas matching real JD requirements, with questions compiled from actual big-tech interviews and common follow-up scenarios, useful for job switchers and new graduates.Does it also teach model fine-tuning?Yes, there is a dedicated section on fine-tuning covering data preparation, LoRA/QLoRA training, LLaMA-Factory practice, evaluation, and multi-GPU/VRAM optimization.
Yes, there is a dedicated section on fine-tuning covering data preparation, LoRA/QLoRA training, LLaMA-Factory practice, evaluation, and multi-GPU/VRAM optimization.What are similar projects?ai-system-design, awesome-chatgpt-zh, ai-guide, ai-engineer-notebooks, ruflo, hello-agents, AgentGuide, awesome-llm-apps
ai-system-design, awesome-chatgpt-zh, ai-guide, ai-engineer-notebooks, ruflo, hello-agents, AgentGuide, awesome-llm-appsWhich topics is it in?coze, deepagents, dify, aigc, a2a, langgraph, lora, langchain, vector-database, curriculum, interview-preparation, vllm
coze, deepagents, dify, aigc, a2a, langgraph, lora, langchain, vector-database, curriculum, interview-preparation, vllm