Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
😮 are you guys aware of brain organoid computers😮 we can make AI out of human br…
ytc_UgxF5Lm4h…
G
If ai gets trained on a model of our own conscienceness it's obvious that it wou…
ytc_UgzCqUr_0…
G
Blah blah vague this vague that blah blah cliche blah the end -> yawn and compl…
ytc_UgwT-mt2e…
G
Would they though? Assuming the worst case scenarios, they'd just be another bun…
ytr_UgzbuTHpQ…
G
As someone who is extremely anti ai art and digital archiving, and as an artist …
ytc_Ugz2rgc_J…
G
So if you lets say use your voice for utube, and then your day job trains some a…
ytc_Ugyq-JnHc…
G
Ai isn’t the best for references since it usually gets alot of stuff wrong (eg p…
ytr_UgzNoJ6M9…
G
No need to check w chatgpt. Bible already confirm Jesus was a messenger of God…
ytc_Ugy8pnVLS…
Comment
AI will make the role of experienced devs more vital than ever before. To properly leverage AI tools you need to have your code base nearly immaculate. Clean encapsulation, clean well thought out interfaces and architecture, detailed project planning and issue descriptions, great documentation and specs, comprehensive build and tooling systems to catch errors and enforce quality, a strong PR culture and the ability to read and review any code (no rubber stamping). I could go on and on.
Overall you just need tons and tons of background knowledge to know the right techniques and solutions to ask for, and you need to be able to understand every line it outputs. It's a common saying that it's harder to review a PR than it is to create one.
Good code is not about the business logic, it's about the architecture, culture, documentation, tooling, extensibility, maintainability, readability, security, scalability. It's creating an entire information architecture system that can be scaled both on an execution level but also an organizational and team work level.
This industry is really inundated with totally out of touch management and pretty amateurish juniors who can only do business logic on a project or framework built by someone higher level than them, and seem to think that's all there is to programming and don't have any desire to go beyond that. We've had to learn the hard way the importance of expertise and experience and applying that to create high code quality code bases. I'm worried that the industry has become so diluted that those lessons have been forgotten, and with AI slop being committed to production the disaster waiting to happen will be bigger than the last time companies axed their teams and destroyed their core competency thinking that their well running systems would just coast without the experts.
reddit
AI Jobs
1745605889.0
♥ 1
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | none |
| Reasoning | mixed |
| Policy | none |
| Emotion | approval |
| Coded at | 2026-04-25T08:33:43.502452 |
Raw LLM Response
[
{"id":"rdc_mp0j5f2","responsibility":"none","reasoning":"mixed","policy":"none","emotion":"approval"},
{"id":"rdc_mp1rpmv","responsibility":"none","reasoning":"mixed","policy":"none","emotion":"indifference"},
{"id":"rdc_mp28qde","responsibility":"none","reasoning":"mixed","policy":"none","emotion":"mixed"},
{"id":"rdc_mp2w6ay","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"approval"},
{"id":"rdc_mp3144e","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"indifference"}
]