If I were mtsears4 - after such reply I would dig a deep hole, hide there and cry for a week.

Dude thought he is smart but ended up being an entitled brat.

*I've worked on teams where multiple engineers argued about the "right" way to build something. I remember thinking that they had biases based on past experiences and assumptions about what mattered. It usually took an outsider to proactively remind them what actually mattered to the business case.*

Gosh I am so tired with that one - someone had a case that burned them in some previous project and now his life mission is to prevent that from happening ever again, and there would be no argument they will take.

Then you get like up to 10 engineers on typical team and team rotation and you end up with all kinds of "we have to do it right because we had to pull all nighter once, 5 years ago" baked in the system.

Not fun part is a lot of business/management people "expect" having perfect solution right away - there are some reasonable ones that understand you need some iteration.

>someone had a case that burned them in some previous project and now his life mission is to prevent that from happening ever again

Isn't that what makes them senior ? If you dont want that behaviour, just hire a bunch of fresh grad.

No, extrapolating from one bad experience to universal approach does not make anyone senior.

There are situations where it applies and situation where it doesn't. Having the experience to see what applies in this new context is what senior (usually) means.

The people I admire most talk a lot more about "risk" than about "right vs. wrong". You can do that thing that caused that all-nighter 5 years ago, it isn't "wrong", but it is risky, and the person who pulled that all-nighter has useful information about that risk. It often makes sense to accept risks, but it's always good to be aware that you're doing so.

It's also important to consider the developers risk tolerance as well. It's all fine and dandy that the project manager is okay with the risk but what if none of the developers are? Or one senior dev is okay with it but the 3 who actually work the on-call queue are not?

I don't get paid extra for after hours incidents (usually we just trade time), so it's well within my purview on when to take on extra risk. Obviously, this is not ideal, but I don't make the on-call rules and my ability to change them is not a factor.

I don't think of this as a project manager's role, but an engineering manager's role. The engineers on the team (especially the senior engineers) should be identifying the risks, and the engineering managers should be deciding whether they are tolerable. That includes risks like "the oncall is awful and morale collapses and everyone quits".

It's certainly the case that there are managers who handle those risks poorly, but that's just bad management.

Nope, not realizing something doesn't apply and not being able to take in arguments is cargo culting not being a senior.

No one wants to spend time writing "I agree", mostly they move on or give an upvote. Doesn't look like TFA counts upvotes as we don't see them.

Somtetimes people write "This", and that's apparently a no-no. You are told to just upvote.

Maybe there should be a setting for hiding such short replies or something like "shadow ban", you can write "thanks" or "This." and only person posting it will see their own "thanks".

Downside is that there is still some cost to it, like writing "please" and "thank you" to LLM...

The counterargument is that, if you think a post is idiotic, you could say so but, if you don't articulate why in detail, you'll probably be downvoted or modded. So better to just downvote if you care and move on.

I guess somewhere this year having your own GPU might be cheaper than renting.

When was buying a device from the manufacturer not cheaper than paying a middleman to manage rentals of said device?

When you need to scale up and down dynamically.

Then the claim above can never be falsified because there will always be a value you can plug into the equation at which it's cheaper to rent (like at 50% occupancy, at 5%, at 0.5%... so long as you're not the only person in the world who wants to use the device being offered for loan). I would assume we're talking about a common workload where one actually needs the device that one considers purchasing for some reasonable fraction of the time and not less than half the time

So in summary:

- own the hardware, if you have constant load and have enough expertise (personally or a team) to use/maintain it at your required reliability level

- rent the cloud, if your usage is opportunistic/ spike’y/rapidly changing

- also rent the cloud, if required expertise/maintenance would cost you more than the hardware (if you want to have gazillion of nines, you need somebody who’s there to deal with smoking stuff, in several locations)

That sounds like good common sense, though I'd reword the "constant" with like <30% occupancy because cloud premiums are rather large

My question was (to what ozim wrote), how's that new in 2026

I wonder what will happen with cloud GPU prices once the AI bubble pops. Will they keep having high prices, or will they drop prices significantly ?

My guess is, the bubble popping means the demand wasn't real. No amount of price reduction will motivate it. Anybody who has to get theirs will be in for a steep premium.

If they do that, that's exactly why you don't want to promote them because it is clear they don't understand that doing x+5 work on your own is not as good as x*5 when you become multiplier by helping others.

x+5 is greater than x*5, if x=1. But great thinking, just make sure you're transparent with your team members so you neither get x+5, nor x*5.

Except the only way to do x*5 work is by your team hiring extra 5 people for you to manage... or, somewhat uniquely to our industry, through automating your own work.

Also, everyone else hears the same memes about "being a force multiplier" too. When everyone is trying to be a multiplier for the team by helping everyone else on it, the result isn't exponential productivity growth - it's drowning in exponential noise.

Like some other commenters correctly observed, the most significant factor is actually whether the company you're in is stable headcount-wise, or growing fast. In a stable company, promos are a contested resource, which makes the requirements arbitrary - you're graded on an ordinal scale, not a nominal one. In a fast-growing company, promos will happen to you, through no effort on your own - you can coast upwards on seniority alone.

In neither situation, consistently performing at the level above you is a differentiating factor.

Getting LLM to write the classifier should be the way to go.

That’s what I mostly do, I give it some examples ask to write code to handle stuff.

I don’t just dump data into LLM and ask for results, mostly because I don’t want to share all the data and I make up examples. But it also is much cheaper as once I have code to process data I don’t have to pay for processing besides what it costs to run that code on my machine.

Why do you need to remember that when you can look it up?

Important part is knowing there are special networks.

Problem is there are people who buy tickets for each and every draw for years.

That amounts for not so trivial amount of money - would be much better for them if they put it in savings account or basically anything else.

Downside is “elsewhere” is going to cut your reach when you post a link because they want their users to stay in their estate.

I believe most of it is people expecting stuff to work differently, not having time to wrap their head around proper usage of system, because they need specific outcome and they don't need mastery of the tool.

Downside is that "Facebookization" created a trend where people expect everything to be obvious and achievable in minimal amount of clicks, without configuring anything.

Now "LLMization" will push the trend forward. If I can make a video with Sora by typing what I want in the box, why would I need to click around or type some arcane configuration for a tool?

I don't think in general it is bad - it is only bad for specialist software where you cannot use that software without deeper understanding, but the expectation is still there.

It is weird to push the idea that Facebook is some kind of pinacle of good and easy to use UI. That's the first one. It's quite the opposite, with people constantly complaining how bad, clunky and confusing Facebook is. And it is not the recent trend either. It has always been this way and e.g. VK has always had a better UI/UX that Facebook (and Telegram's is better that Whatsapp's).

Not as pinnacle of good - but pinnacle of “you don’t have to think, just scroll, occasionally like something” ;).

Then people expect accounting software to be just login click one or two buttons.

But still, compared to something like email, the previous standard for most people, Facebook was an unbelievable step forward. People complain about anything.

Facebook is a step forward in terms of features. But it is a clear regression in terms of UI, easiness, usability, understandability of email. Email is very simple in both concept and practice.

I think I disagree - when it comes to sharing large files, something like a video, or even a picture in 2005, email was nowhere near as good. And also having a place to comment on things without the stacking up of reply chrome is genuinely better.

Email was created for email, not for file sharing of massive files (even though nothing stops). Facebook is even worse than email at sharing large files, video files or even pictures (like, imgur is much better).

Commenting on things is from a list of features (to be distinguished from UX/UI) I talked about.

> when it comes to sharing large files, something like a video, or even a picture in 2005, email was nowhere near as good.

That's just a stupid limitation and not even a technical one. You could happily send GBs over email. You can also easily filter allowed attachment size by sender on the recipient side, because by the time the attachment size is told, both information was already provided.

facebook literally obfuscates their UI to stop you turning off "features" they want to push on you

it is a UI designed to be hard to use

You're talking about two different things here (and I'm not condoning either, to be clear.)

1) UI = a clearly documented way to configure all features and make the software work exactly how you want.

2) UI = load a web page and try to do the thing you wanted to do (in this case communicate with some specific people).

FB is clearly terrible at 1 but pretty alright at 2.

> If I can make a video with Sora by typing what I want in the box

IME, people cannot even articulate what they want when the know what they want, let alone when they don’t even understand what they want in the first place.