Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
Not the problem that people make it out to be. We still thankfully have 8 billio…
rdc_k9hrrvq
G
This is a psychological operation in that they need to give the AI robot life an…
ytc_UgxNqnFkR…
G
You know what ChatGPT and other text genAI is good at? Rephrasing a limited amou…
ytc_UgwsYBToX…
G
He have speziel ChatGPT 😂😂😂😂 my ChatGPT says it’s easy possible for humans or AI…
ytc_UgwcmMpsh…
G
AI is a tool. It is a tool easily misused. It has uses, even for images.
Like ge…
ytc_UgyE89DLv…
G
And AI "Artist" calling themselves an artist would be like me calling myself a c…
ytc_UgzDF-p_t…
G
Maybe AI will convince us how to eat healthy which will finally motivate most of…
ytr_Ugw2wMG7H…
G
i genuinely hope ai bros continue to get and stay mad. this is a really well wri…
ytc_UgztcStGu…
Comment
>Did you regret joining google?
Nope. I think I had to learn what this kind of friction really is to learn what I really want out of a job. Also, the benefits and pay are literally the best in the industry, so it was a very *safe* place to learn this lesson.
>Would you have joined if you knew it was like that.
Yeah, I'm pretty egotistical so I thought I'd be able to make some small change and prove myself. I didn't realize how much momentum is in the status quo.
>Also I’m curious did google do pull request reviews or code reviews?
Yes. It's a tiny bit of an annoying part of it. From my open source work, we used github and did pretty standard PR and code review stuff. But on the internal monorepo, most of my work there involved other teams, so I needed to get reviews from someone on that team, and someone with readability certification. This conceptually is very good. But often they don't have enough context to understand my feature, so they often can't help stopping me from footgunning myself.
There is also a lot of automation, so you often have to fight the linters and tests. Which conceptually sounds like a good thing, but they're often vague and not super helpful to understand what the problem is or how to fix it. Other times, there is an automated tool that will fix it for you, and I wonder why do I need to run this manually every time?
reddit
AI Jobs
1747929341.0
♥ 15
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | none |
| Reasoning | unclear |
| Policy | none |
| Emotion | approval |
| Coded at | 2026-04-25T08:33:43.502452 |
Raw LLM Response
[
{"id":"rdc_ohqqfl2","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"outrage"},
{"id":"rdc_ohqssdy","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"outrage"},
{"id":"rdc_ohqwk0t","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"outrage"},
{"id":"rdc_oi373ng","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"rdc_mtoawyx","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"approval"}
]