Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
My bachelors degree in commercial saxophone performance is safe and intact. Can’…
ytc_Ugw81UrVW…
G
I work in Tech and I know people who can't even prompt an LLM. We need to assess…
ytc_UgwthPlEf…
G
@UnHysteria If it looks good, then is not slop, the only difference is that one…
ytr_Ugw9QY8LU…
G
That's the entire point, to us it looks like nothing changed, but to an AI progr…
ytr_UgwDNwXfu…
G
100%, this is disheartening. Control AI is sponsoring creators for a reason, and…
ytr_Ugx06C4bE…
G
Nobody wants this. It’s not gonna happen. The rich people can dream all they wan…
ytc_Ugzl4eYM4…
G
Why would you do that..because in the long run the racism will still be programm…
ytc_Ugz6STbsf…
G
for most of the time, at least so far, we hear about these concerns in the weste…
ytc_UgydmIFtJ…
Comment
I haven't been using Claude for many transformation tasks. I've used it and ChatGPT for some file transformations before (like converting a TS configuration file into a JSON) and it did pretty well, but you'd probably be better off having the model write you a function that will do the conversion rather than relying on the LLM to output the correct conversion if you're doing trivial file format conversions. Many times in the past I've had transformations missing some data. It's just not really what the tools are made for as far as I can tell, at least in cases where you require a lot of data to be converted
Anyways, since you asked, I'm a modern TS React and Node developer. I also use the AWS CDK for infrastructure management in TS. Claude is insanely, unbelievably good at writing in modern TS code if you understand how to design your prompt well, if you have a good understanding of your codebase, and aren't using niche libraries. I will occasionally write a few lines of code these days, but 99% of it is AI generated. Even simple tasks like implementing deadlock retry handling in DB queries, or error message handling in some React components, it makes much more sense to delegate to AI than to write yourself. I wouldn't expect Claude to be as good in a language that isn't strongly typed, or when using libraries that are older / more legacy. I actually wound up migrating to TS in my React project because 1. Vanilla JS dynamic typing is an absolute nightmare, and 2. Having interfaces and types provides AI the same benefits it provides devs, except AI is about 1000000000x faster at contextualizing files than humans
AI code generation isn't "vibe coding", btw. Vibe coding is just dumping AI slop into your project with basic prompts that non-developers could generate. Professional developers who use AI well are designing prompts that require underlying software knowledge and are reviewing the generated code to make sure it makes sense and is compatible with long-term maintainability
Keep giving these tools a try. Maybe your use case wasn't great, but trust me when I say there is hype for a reason. These tools have been an absolute game changer for me and I'm so much more productive I can't even quantify it, it's literally like rocket fuel for modern web app development.
youtube
Viral AI Reaction
2026-03-07T01:4…
♥ 2
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | none |
| Reasoning | unclear |
| Policy | none |
| Emotion | indifference |
| Coded at | 2026-04-27T06:24:59.937377 |
Raw LLM Response
[
{"id":"ytr_UgyzdMFHbjXYBvvt2yB4AaABAg.ATz_znFeccZAU1gFtyrw_9","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"ytr_UgyzdMFHbjXYBvvt2yB4AaABAg.ATz_znFeccZAVv5JAh8X1Y","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"resignation"},
{"id":"ytr_Ugy2s4W_dbkrGUTMpkJ4AaABAg.ATzYG6is8U3AUHZqE7nYtx","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"outrage"},
{"id":"ytr_Ugxn5GEO5cjh_IIzvRh4AaABAg.ATzA3u2-XaPATzY63wTpZH","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"ytr_UgxXsvrDaRMDIehoh9N4AaABAg.ATz7jogCNmOAU-5lbU4RRO","responsibility":"company","reasoning":"consequentialist","policy":"liability","emotion":"fear"},
{"id":"ytr_UgxXsvrDaRMDIehoh9N4AaABAg.ATz7jogCNmOAU-O4j4FQo9","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"approval"},
{"id":"ytr_UgxXsvrDaRMDIehoh9N4AaABAg.ATz7jogCNmOAU05DJJb6Ud","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"approval"},
{"id":"ytr_UgzgXLYuBQQHuQirxiF4AaABAg.ATz78Ay6nLaATz7jIc_6Ez","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"resignation"},
{"id":"ytr_UgznoRo-YBGZnkuG_2R4AaABAg.ATz4JFOW8AaATzB3iXLabq","responsibility":"distributed","reasoning":"consequentialist","policy":"none","emotion":"fear"},
{"id":"ytr_Ugx9wzWDwHLRTu4s3IR4AaABAg.ATz4AU_sWJsAU0o7L6XE3U","responsibility":"user","reasoning":"consequentialist","policy":"none","emotion":"indifference"}
]