ai-engineering-hub vs CodexGuide
36k-star hub of runnable AI engineering tutorials — LLMs, RAG and agent apps as self-contained projects, including build-code-harness: a Claude-Code-style coding harness rebuilt on CrewAI + E2B. — versus — Community-maintained Chinese practice guide to OpenAI Codex — learning paths, CLI/App/Cloud/IDE setup, AGENTS.md templates, sandbox/approval safety and team playbooks, published at codexguide.ai.
Both are learning material rather than tools: ai-engineering-hub ships runnable tutorials for building LLM and agent apps, codexguide is a Chinese-first practice guide for operating one harness â setup, AGENTS.md rules, approval boundaries, team playbooks.
| ai-engineering-hub | CodexGuide | |
|---|---|---|
| Stars | 37k | 3.3k |
| Forks | 6.1k | 311 |
| Language | Jupyter Notebook | PowerShell |
| License | MIT | MIT |
| Last activity | 28 days ago | 4 days ago |
| Topics | coding | coding |
| Curated connections | 3 | 2 |
ai-engineering-hub — the curator's take
A tutorial hub, not a tool — each folder is a complete, runnable project with real dependencies. The standout is build-code-harness: it rebuilds a coding-agent harness (planning, memory, checkpoints, sandbox, human gate) one layer at a time on CrewAI + E2B, the best way to understand what harnesses like Claude Code actually do. Content repo caveat: quality varies by folder, and there's a newsletter funnel attached.
CodexGuide — the curator's take
Use it to onboard people — especially Chinese-speaking teams — onto Codex without the trial-and-error phase: real task flows (PPT, Obsidian, CI fixes, Feishu/Notion), AGENTS.md rule templates, sandbox and approval boundaries, and a team playbook for turning one successful run into reusable process. NOT a tool: nothing to install, it's a VuePress knowledge base. Content is Chinese-first (an English README mirror exists), the README carries a heavy sponsor block, and anything time-sensitive (pricing, availability) should be re-checked against OpenAI's own docs — the guide itself says so.