I learned Odin
TL;DR
Odin won him over fast: Rewriting his old Love2D tower defense in Odin got him to a working level renderer with JSON-loaded data, pathing, and an auto-generated Perlin-noise background after roughly 20 to 30 hours in the language.
Raylib beat Love2D for his use case: He gushes over Raylib's texture and source-rectangle model because scaling, flipping, and drawing feel simpler and easier to reason about than Love2D's interfaces.
Explicit function overloading is his favorite surprise: Odin's version of overloading clicked for him because you must explicitly declare which functions share a name, which made helpers like
cell_indexfeel clean instead of magical.Odin's
usingfeature made grid code pleasant: By embedding coordinates into aCell, he can pass aCellwhere aCoordinateis expected, which fits a row-and-column tower defense game nicely.AI worked fine with Odin, but not magically: He says models did not get confused by the language and quickly generated JSON-loading code, though he still had to refactor heavily because the output was often bad.
The real point is recovering his love of coding: He says the past 6 to 12 months of pushing into AI programming drained his passion, and this dev log channel is his attempt to return to building things by hand because that is what got him into programming in the first place.
The Breakdown
After 20 to 30 hours with Odin, ThePrimeagen says it was "extremely easy" to pick up, good enough to get his tower defense rewrite rendering level data, pathing, and a Perlin-noise background fast. The bigger reveal is why he is doing it: after months of chasing AI coding promises and feeling like "the co-pilot," he wants to get back to programming for the joy of it.
Was This Useful?
Share
Keep Reading
Make Alcreon Yours
Tune your feedFive quick questions, and the feed ranks what matters to you first.Or just get notified
The weekly Echo. Signal worth keeping in your inbox.
Every new piece, announced on X.
Read Next
See all
Playbook
Tasteful Skills
“Tasteful Skills” argues that the best agent skills are not documentation or best-practice lists.

Playbook
The Art of Tasteful Prompting
Learn how tasteful prompting helps you move beyond generic AI output by shaping context, style, and judgment from the start.

Playbook
The Codex /goal Playbook
OpenAI shipped /goal for the Codex CLI. It turns a prompt into a persisted, self-continuing contract.