Raw LLM Responses

Inspect the exact model output for any coded comment.

Comment
not sure I agree with this. I find it most helpful for things that are easy (or at least, easy to google), but tedious. like spawning async web requests, parsing json, reading/writing to s3, manipulating local paths, etc. i've had a lot of luck in one-shotting data parsing/wrangling/pulling scripts with LLMs. i've had to make the prompt very detailed to get clean code back, but spending \~10 mins writing a prompt and then \~5-10 mins reviewing the code is still way faster than writing the script from scratch.
reddit AI Jobs 1744260217.0 ♥ 1
Coding Result
DimensionValue
Responsibilitynone
Reasoningunclear
Policynone
Emotionapproval
Coded at2026-04-25T08:33:43.502452
Raw LLM Response
[{"id":"rdc_mle5pcu","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"approval"}, {"id":"rdc_mmc72p4","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"approval"}, {"id":"rdc_mle6efo","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"indifference"}, {"id":"rdc_mlflhnp","responsibility":"developer","reasoning":"virtue","policy":"none","emotion":"approval"}, {"id":"rdc_nxutfzk","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"outrage"}]