Developer Track
How do I build an agent and ship it
From your first LLM API call to building your own multi-agent framework and evaluation harness. Code, architecture diagrams, and numbers we measured ourselves.
- Readable lessons
- 67lessons
- In progress
- 0lessons
- Est. hours
- 54hours
- Stages
- 4
After this stage you can:You know the vocabulary, what it can do, and have run it once
- 01What Is an AI Agent: From ChatGPT to AI That Acts
用 LLM、上下文、工具、循環四個零件拆解 AI Agent 的本質,對比聊天機器人與 Agent 的差別,給出日常、開發、商業三個例子與真實失敗模式,幫你判斷自己的需求該不該用 Agent。
11 minChinese only - 02Set Up Your LLM Dev Environment in Five Minutes
從零搭好能呼叫 LLM API 的開發環境:Python 安裝、虛擬環境、API Key 取得與環境變數管理,最後用一條 curl 指令驗證整條鏈路。macOS 與 Windows 雙平台完整指令,重點講解金鑰安全管理與不洩漏進 git 的做法。
15 min1 prerequisiteChinese only - 03Your First LLM API Call: Tokens, Billing, and Common Errors
用 Python 的 openai SDK 發出第一個 LLM 請求:搞懂 system、user、assistant 三種訊息角色的分工,temperature、max_tokens 與串流輸出的取捨,以及 token 計費概念與用量估算方法,最後整理新手必遇錯誤的排查步驟。
21 min1 prerequisiteChinese only - 04Understanding the Transformer Architecture: Attention Is All You Need
An in-depth understanding of the core components of Transformer: Self-Attention, Multi-Head Attention, Positional Encoding, and how they replaced RNNs to become the foundational architecture of modern LLMs.
7 min1 prerequisite - 05Prompt Engineering Systematic Methodology: From Few-shot to ReAct
A complete methodology for Prompt Engineering: principles and practical applications of techniques such as Few-shot, Chain-of-Thought, Tree-of-Thought, ReAct, and Self-Consistency.
6 min1 prerequisite - 06May 2026 LLM API Pricing Landscape: Complete Comparison of DeepSeek, Qwen, GLM, Kimi, MiniMax, and Doubao
Cross-verified through multiple search engines including Exa, Tavily, and Brave, this compiles per-million-token pricing for DeepSeek V4, Qwen3.6-plus, GLM-5, Kimi K2.5, MiniMax M2.7, Volcano Engine Doubao Seed 2.0, and Claude/GPT, with a Sub2API cost setup guide.
12 min1 prerequisite - 07Three-Layer Framework LLM API Configuration Summary: DS V4 Pro, Qwen 3.6, oMLX Local
A complete guide to configuring DeepSeek V4 Pro, Qwen 3.6-Plus, and oMLX local Qwen3 Coder Next across Hermes Agent, OpenClaw, and Claude Code.
9 min1 prerequisite - 08Highly Useful Claude Code System Prompt Templates
A curated set of battle-tested Claude Code system prompts, covering common scenarios such as code review, refactoring, and documentation generation.
5 min1 prerequisite - 09Your First Agent: A Hello World That Uses Tools
不用任何框架,手寫一個真的會做事的 Agent:會查天氣、會算數學、自己循環到任務完成為止。逐步拆解 Agent 循環的六個步驟、終止條件與常見踩坑,寫完你就懂框架替你做了什麼。
28 min1 prerequisiteChinese only
Switch track
The tracks share the same articles, so progress here is not wasted — switching only changes the ordering and the extra lessons.
How much time can AI save my team, and how do we roll it out
How do I use AI for research, due diligence, and valuation — without being lied to
Curriculum structure last updated 2026-09-29. Content is still being filled in; lessons marked “in progress” are not live yet.