Nate Combs used ChatGPT for research and a prompt, and Claude Code implemented the expansion, iterating on his playtest feedback.
Nate Combs used ChatGPT for research and a prompt, and Claude Code implemented the expansion, iterating on his playtest feedback.
ChatGPT did the historical research on the battle of Alesia and wrote the prompt for Claude Code. Claude Code then built the expansion for a remake of a 1990s video game. The map was reconstructed from text sources ChatGPT supplied, including a Caesar translation, The Latin Library, Britannica and a French archaeology museum. Nate Combs playtests it and gives criticism, and Claude Code revises.
Combs calls the expansion odd but gripping and shares screenshots. He was still playtesting when he posted, so there is no final result or measured outcome.
The source gives no cost or time figures. The workflow depended on ChatGPT's research and prompt handing off to Claude Code, plus repeated rounds of playtest feedback.]
I have the source for my remake of [classic game] in [project path]. Add an expansion scenario based on [historical battle, e.g. the Battle of Alesia] using this research brief: [paste notes]. Implement the map, units, and win conditions, then let me playtest and iterate on my feedback until I say it's done. Keep changes on a separate branch.
Claude Code got an old DirectX game running on modern hardware, decompiled much of it, built a headless parallel runner, and now writes full level guides with screenshots.
Plan, execute and review a cross-cutting refactor in small batches, with Claude Code making the edits and GitHub Copilot reviewing each PR.
Claude Code reads your coverage report, picks the riskiest untested modules and writes unit tests that pass, one small PR at a time.