Benchmarking AI agents across five TypeScript back end frameworks

Interesting, agents seem to always fall in a spectrum between overengineering and underengineering, where they will either go wild and overengineer a simple solution, or do the minimal effective thing to pass with "//TODO fix for prod".

It's the same patterns we see with human devs, just applied on codebases at scale.

The conclusion has an interesting tibid tho - maybe the frameworks and their developers should start including more abstract focused primitives instead of just the low level ones, similar to what Encore did, as that way the behavior is encoded at the framework level.