Home · ← Claude & Claude Code | 2/8 | VS Code →

ChatGPT & Codex

The parallel track — an alternative reasoning LLM and agentic coder

Overview

ChatGPT & Codex

Everywhere the course says "Claude / ChatGPT," this is the other half of that choice — not a lesser option, a genuine alternative.

What it is

What it is

  • ChatGPT (chatgpt.com) — chat reasoning LLM with a built-in Data Analysis tool.
  • Codex — OpenAI's agentic coding CLI, the direct alternative to Claude Code. It reads AGENTS.md too.
Where it fits

Where it fits

D1·1Environment check
D1·2AGENTS.md
D1·3Literature tools
D1·4Referee skill
D1·5Subagents
D2·1Hypothesis
D2·2Design
D2·3Data lab
D2·4Results
D2·5Manuscript
D2·6Governance

Offered as an alternative to Claude/Claude Code from setup through referee — not used for the Claude-Code-specific skill/subagent exercises (D1·2–D1·5).

Access & cost

Access & cost

  • Sign up at chatgpt.com.
  • Plus plan (≈ €23/mo incl. VAT) includes Codex and Data Analysis.
In the course

D1·1 — an alternative install

The tool setup lab installs “Claude Code / Codex” side by side — pick one or try both. The same AGENTS.md ground rules apply either way.

In the course

D2·1 & D2·2 — Hypothesis & design

The schedule lists “Claude/ChatGPT, optional Gemini” for the hypothesis loop, and “Claude/ChatGPT” for design & identification. The generate–critique–refine loop and the adversarial design review work the same way with either model.

In the course

D2·3 & D2·4 — Data, results & robustness

“ChatGPT Data Analysis” is the listed alternative to Claude Code for running the analysis and finding the pre-trend confound (D2·3), and for writing up the naive vs. robust estimate (D2·4).

In the course

D2·5 — Manuscript

Paired with Paperpal/Writefull in the schedule (“Claude/ChatGPT + Paperpal/Writefull”) for the writing session — ChatGPT drafts, a dedicated writing tool polishes.

In the course

D2·6 — Referee

Works as an alternative critical-reviewer persona for the referee pass, same checklist logic (causal language, pre-trends, citations, numbers, disclosure) run through a different model.

Why it's here

Why have two models at all

A design memo or referee pass reviewed by only one model can share that model's blind spots. Running the same adversarial question through both products is the cross-tool version of D1·5's “do two independent checks agree” pattern — applied across products, not just subagents.

Limits

What it doesn't replace

The course's rules — no invented citations, no bare causal claims, name the identification strategy, disclose every AI-assisted step — aren't features of any one product. Your own AGENTS.md for ChatGPT/Codex should say the same things Claude's does.

Recap

Recap

See the home page for the full pipeline — every stage where this tool appears is listed as a Claude/ChatGPT choice, never ChatGPT-only.