Raw LLM Responses

Inspect the exact model output for any coded comment.

Comment
If the limiting factor to your company's productivity is how quickly engineers can type their code, AI sounds like it would work great. But in the real world, most engineers I see spend most of their "coding" time trying to understand the problem, architect a solution, and develop strategies for testing and debugging that solution. Unless ChatGPT has come a long way since I last used it, it's more or less useless for doing these things.
reddit AI Jobs 1687884228.0 ♥ 944
Coding Result
DimensionValue
Responsibilitycompany
Reasoningconsequentialist
Policyindustry_self
Emotionindifference
Coded at2026-04-25T08:33:43.502452
Raw LLM Response
[ {"id":"rdc_lngelzu","responsibility":"none","reasoning":"contractualist","policy":"regulate","emotion":"approval"}, {"id":"rdc_lnixapx","responsibility":"none","reasoning":"deontological","policy":"liability","emotion":"outrage"}, {"id":"rdc_lnkjtgg","responsibility":"none","reasoning":"deontological","policy":"liability","emotion":"outrage"}, {"id":"rdc_lnnj80y","responsibility":"none","reasoning":"deontological","policy":"none","emotion":"outrage"}, {"id":"rdc_jpqojex","responsibility":"company","reasoning":"consequentialist","policy":"industry_self","emotion":"indifference"} ]