anon22981

Honest take: I work less now with maybe slightly faster results while the output quality is probably about the same. Some bugs sometimes fall through, but when hasn’t that been the case?

I take effort to write specs for the agent and I take effort reading PR’s but thats maybe an hour of work. Obviously there are meetings, sometimes tasks require more planning, sometimes I investigate production problems/data more manually, but overall - I feel I work less after agents became a thing. Kind of ashamed of it tbh.

I feel the 1M context is way too large —- the model gets ”drunk” way before it gets anywhere near 1 million. Imo the 1M context window is a huge downgrade.

I suspect people with opinions like the author’s haven’t been in a project where people use LLMs responsibly. We had a senior dev basically just prompt and push, with very little overseeing and minimal instructions, causing so many bad PRs and even prod bugs. That made me a sceptic for many months about agents.

Then we started to have (myself included) people actual plan out the tasks for the bot: give good specs, good ac, file context, better self-review, better ”agentic practices” (i.e. asking it to review it own work can sometimes help), and suddenly I noticed you really can use agents in a real world 1mil LOC project. If you do it well and responsibly (also meaning you still retain some sense of ownership and actually review the shit)

I still write my cover letters myself. :( Though I do refine them with LLMs, but instruct to retain the tone and the feel of the writing.

I really like going to linkedin daily to play minisudoku and a couple of other puzzles, then never engage the feed or other features

I use LLM’s daily and agents occasionally. They are useful, but there is no need to move any goal posts; they easily do shit work still in 2026.

All my coworkers use agents extensively in the backend and the amount of shit code, bad tests and bugs has skyrocketed.

Couple that with a domain (medicine) where our customer in some cases needs to validate the application’s behaviour extensively and it’s a fucking disaster —- very expensive iteration instead of doing it well upfront.

This. I’d move to Linux in a heartbeat if certain anticheats for certain competetive games had supports for it. (i.e. faceit anticheat)

What I’ve seen also happen is senior devs suddenly starting to put out garbage code and PRs. One senior dev in our project has become a menace and the quality of his work has dramatically dropped.

Sounds like a junior vibe coder with no understanding of software development trying to boost their CV. Or at least I hope that’s the case.

Those lazy employees need that strict supervision!

Maybe these c suites and other employee hating assholes are projecting their own lazyness. Or maybe they think they are so superior compared to ”common” people that the ”common” people must be lazy trash.

I don’t know, but it is weird to assume most people won’t do their job without ”strict supervision”. Like super weird.

(Btw, anecdotally, most people I know work more efficiently from home with fewer breaks)

Since you are only willing to go for 1:1 odds with a 3 year timeframe, I assume you are in agreement that it might happen? Otherwise I’m sure you would give him better odds with a larger timeframe :)

Mikko Hyppönen, who holds at least some level of authority on the subject, just recently said in an interview that he believes currently the defenders have the advantage. He claimed there’s currently zero known large incidents where the attackers have been known to utilize LLMs. (Apart from social hacking.)

To be fair, he also said that the defenders having the advantage is going to change.

Tesla definitely has suffered a huge blow to it’s reputation (or several of them) which is evident in the EU new car sales.

Maybe your take is true if you are non-EU, but it definitely isn’t true in the slightest in Europe and is pure copium.

To be perfectly fair saying ”it’s aware of the best practices, context and internal usage” is very misleading. It’s aware none of those (as it is not ”aware” of anything), and that is perfectly clear when it produces nonsensical results. Often the results are fine, but I see nonsensical results often enough in my more LLM dependant coworkers’ PRs.

I’m not saying not to use them, but you putting it like that is very dishonest and doesn’t represent the actual reality of it. It doesn’t serve anyone but the vendors to be a shill about LLMs.

How to reveal your multi-layered ignorance and delusions in a single, short comment. Impressive in a way.

Just for anyone unfamiliar with the game or these types of games and to put out an opposing view — I disagree that the experience of BR games have anything to do with the experience of gambling. I’ve done both and find pretty much nothing in common.

(The cosmetic gambling obviously is very much gambling, though, and is unfortunatly a big part of the game.)

Do you (or anyone!) know a good tutorial / series / blog how to get started with home automation?

I can’t speak for the lower ranks, but I was mid-ish on premium this season (played from 17k to 23k then stopped iirc) but a lot of the cheaters were subtle enough that not everyone realized that they were cheating, even though they pretty blatant from my perspective. (For reference, I play lvl 10 faceit and have a decent understanding of the game.)

Point being that a ”subtle hacker” might be subtle to some and obvious to others. So OP, being a newbie with only 6 mo of experience, might suspect someone being subtle about their hacking but they might be very blatant to a more experienced player.

My thoughts exactly reading these comments. Generating scripts is a very good use for LLMs, but I was under the impression that vibe coding generally referred to building whole products with just prompts. It’s a very different beast to do a small script (or a todo app) than to do a relatively more complex application.

Instead of being rude and snarky you could just answer the question. Or just not reply at all.

Not everyone has looked into or tried everything.

They are being silly, I think it’s free for download on the internet

Yeah, my and my SO have very different experiences regarding this.

I’ve quit cold turkey several times with little problems - quit for several months or over a year, then used again for x months, then quit again, etc. (This went hand in hand with a certain seasonal job.) Apart from very slight ”I wouldn’t mind a smoke” kind of feelings it doesn’t have much of an effect of me.

My SO was miserable, depressed, craved cigarettes for months and months. For the longest time she was convinced that it isn’t possible for her to quit, since they cravings didn’t go away. (Until they did.)

I’m not sure if you are trolling, but I’m pretty sure just attaching a file to an email is often more efficient. Not to speak of legibility or maybe the intended use for said documents. Sadly I don’t think you are trolling.

Also: a computer (where the documents were) has the required hardware and software to send the documents as is. Your phone is the special and external hardware in this case, not the computer (also probably more expensive than the computer). Your phone requires signups and logins as much or even more than computers. I don’t know what paid subscription you would need to share these documents. Plenty of free methods. I don’t know why you are talking about malware? Do you think the documents might mutate into malware while traveling the internet? I think the risk exists with pictures too, be careful out there!

I disagree strongly on the points of batteries and cables, and so would most of people who care about gaming mice to begin with (competetive gamers). Same probably holds true for most casual gamers as well.

To be honest, your whole comment seems to be very out of touch when it comes to modern PC gaming.

Yeah nah. The endless loop of useless suggestions or ”solutions” is very easily achiavable and common, at least on my use cases, not matter how much you iterate with it. Iterating gets counter-productive pretty fast, imo. (Using 4o).

What do you mean it’s before ”vibe shift”, that’s not true?

Not the US but most EV markets, apart from maybe eventually asian markets. Tesla and Musk had a crazy amount of goodwill and pretty much everyone I know was ”oh look, it’s a Tesla!” early on. Now every is more like ”oh look, it’s a Tesla..”

The point is still kind of valid, it is a decently affordable EV compared to many other ”luxury” EVs. They also had (have?) very cheap payment options available. I’d suspect the expenses are a big factor in the purchasing decision.

edit: obviously it’s not a car for poor people as it is still very expensive. I understand your point too.

Seems more like trying to find out the most proficient LMM users than anything else. I’ve never done interviews but I imagine I’d be hard pressed to skip candidates solely because they aren’t using LLMs.

Each to their own and maybe their method works out, but it does seem whack.

I find it laughable this whole thread is people throwing around opinions instead citations