youtube.com
Multiple parts in the workflow stuck out to me:
- "Pair programming" felt a lot more like jamming on ideas than anything
- Copy-pasted real government data into ChatGPT and generated JSON that would be the backbone of the game. Just gave websim that file and copy-pasted the schema, and that was enough to make the bare bones of the project.
- Morry's prompts were super detailed and specific
- When things weren't working, reverting to an earlier version where they did work and writing a different prompt often worked.
- Prompted Claude with a screenshot of the DOGE website, which totally transformed the visual style of Doge Decimator.
- The hardest part of vibe coding a game is to actually think big picture about what the game loop is. Most people vibe code toys and playgrounds, not games.
These outcomes are interesting, they feel diverse. The level of experience you can craft with just a little bit of time and patience is insane.
Watching along in the background as people make things, as this was made, it's unique to say the least. It's taught me a lot and opened up so many doors on a creative level.
Simply fascinating in every way.
for example, saying "let's do it with a tinder ui"... could have been more than enough to visualize in your mind or do a rough drawing... and then just importing one of the dozen tinder clones ui for react... instead you wasted freaking 94 prompt iterations just to correctly visualize what you were think in the first place. and probably accepted the agent small decisions which were probably not your first choice just because you got prompt fatigue.