Raw LLM Responses

Inspect the exact model output for any coded comment.

Comment
It’s overhyped but it’s also worth mentioning that your job is particularly an outlier in terms of how ineffective an LLM would be. LLMs need large amounts of existing codebases/work to make “inferences” and predictions from based on a given input. If you’re writing basic CRUD apps in Javascript and/or a popular javascript framework… LLMs can feel like magic when prompted for rudimentary junior engineer work. Outside of that, they still struggle. It’s a better tool to iterate with and pretend they’re a coworker. Bounce ideas off. I like to prompt and then browse google at the same time and review the output.
reddit AI Jobs 1743885139.0 ♥ 1
Coding Result
DimensionValue
Responsibilitynone
Reasoningconsequentialist
Policynone
Emotionmixed
Coded at2026-04-25T08:33:43.502452
Raw LLM Response
[ {"id":"rdc_mlgwbx1","responsibility":"none","reasoning":"mixed","policy":"none","emotion":"approval"}, {"id":"rdc_mlgymje","responsibility":"none","reasoning":"mixed","policy":"none","emotion":"mixed"}, {"id":"rdc_mlip6q9","responsibility":"none","reasoning":"mixed","policy":"none","emotion":"indifference"}, {"id":"rdc_mllfdrm","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"mixed"}, {"id":"rdc_mlmlkhq","responsibility":"none","reasoning":"mixed","policy":"none","emotion":"approval"} ]