Next step on my job hunt... Automated coding interview... All the surveillance bullshit. Share your whole screen, camera and mic, we will record every word. Don't look anything up, no AI.

Fine.. Can do, you surveillance state arseholes... So I get to coding. Coding challenge stuff. Easy enough. Took me ten minutes to get into the groove. Then I made it in optimal big O. Some smaller stupid mistakes. Not incorrect.

But I have a toddler. Anyway, turns out I sing to myself when I am focusing. So, when they open. The video, they will find the applicant solve their test while singing:

Twinkle, twinkle, little star,
How I want to know what you are,
Up above the world so high,
Like a diamond in the sky...

Guess next time I also should try not to look like a psychopath... Let's see I'd someone will bring it up.

Don't you find it annoying when the dev team gives you the worst or less than absolute minimum of time to test a new release?

It's like... that itches with me so hard. Come on.. we're testing a build here. You want me to brush over this? The amount of bugs that might pop up...(oh yeah we just bill the client for those later lolololol)! Especially when no one wrote tests, or just minimal, I... sigh.

lol

I am job hunting atm.

I fucking hate it. Just sent half of my life to them with lied niceties. Got dammit... Data farming was never simpler.

All I need to do is set up a fake company website, put out a few job ads and let people send me their certificates, cv, telephone numbers. Guess those could be sold.

Hey kids, be careful what you're telling on the internet, except if you want a job, then run around naked.

Oh, yea.. shameless plug.. If anyone needs a web developer, full stack, anywhere in Europe where there is a beach and a warm climate, who pays >90k €. You'll get a great developer, who is perpetually in a bad mood, has a never ending headache, is blunt and does freelance work on the side.

So, anyone? No one? Yea, didn't expect so.. Back to lying to get a job. Indeed, here I come.

Shortly after joining the $company, I found this issue where if you connect a certain peripheral to the console, you can use it to do weird stuff that isn't really a thing on consoles. I documented my finding by creating a bug report.
...the bug report had been closed by the test lead w/ a message stating 'The game doesn't support the peripheral. Won't Fix'.
Fucking moron. Well... personally I don't give a fuck whether the issue is fixed. For one, though, it's !that tester's call - it's the devs' call. Seconds thing about this is - if the peripheral is !supported, then how about fucking disabling it, instead of letting the game intercept input from it, which can /* && does, as it turns out */ lead to unintended behavior?
I kind of wish the game had an SQL injection vulnerability, so I could create a report of it just to see whether the retard closes that one too w/ 'The game doesn't support code injection. Won't Fix'.
FFS.

I got hired in as a senior engineer and, after getting adjusted to the new job, I suggested some pretty big process changes for a big year long project we're working on.

I convinced the manager and team to not continue our practice of dozens of independent project repos where we repeat a lot of the same .NET code, and instead have all our services in one big monolithic repo specific to this domain. I build out the docker compose infrastructure so we could build everything using local Kafka and Mongo (previously devs always relied on the cloud dev environments for both).

It's streamlined, and still a bit clunky because I didn't want to change too much, but I think it will be a significant gain with a project that's this big.

We're already seeing a lot of issues that could have been swept under with the old process of tons of independent services.

Still, I struggle to care. We had layoffs in November, the logistics industry is still struggling, my manager is still using LLMs for stories and coding .. and although he did fix a bug in our Kakfa libraries pretty quickly using copilot, elsewhere it just seems like we're getting big messes of stuff. Even I'm auto-generating all the unit tests and not bothering to even look at them anymore.

I've been here six months. I bet I'll see this go into production. I hope it's not a disaster. I'm struggling to care.

Looks like devrant paid for their certs and https is back on the table (so I am back).

Update: I signed an offer two days ago. I battled myself for a while because I am self conscious when it comes to money, but talking to the new team for 3+ hours and chatting with some close peers helped me realize that I have enough money, it's still good money I'd be making, and my work life balance would be great even for a startup.

Saying the job market is rough would be an understatement. It's really a battle of perseverance and a numbers game.

When I applied for the AWS re:Invent 2025 All Builders Welcome Grant, I honestly didn’t expect to get selected.

But a single email changed everything.

In this video, I share my real journey of getting selected for the AWS re:Invent All Builders Welcome Grant 2025, including how I discovered the opportunity, what AWS actually looks for in the application, and why community involvement and networking matter more than most people realize.

https://youtube.com/watch/...

devRant is the ultimate backend to test my proxy server with since it's random responses. When your proxy server can handle devRant scenario's, it's robust. Making a proxy server that is actually robust and don't let it flaws itself with slow connections / servers is quite a challenge. Depends also what you're writing it with. In python was way simpler than in C. In Python some things magically just worked. If you think that python sockets are native, you're wrong kiddo, they did al sort of shits under the hood. Had to do a whole rewrite of my original proxy server to allow pipelining. Pipelining is req>req>req>resp>resp>resp instead of the traditional req>resp>req>resp>req>resp. It's keep-alive on steroids. Most server don't support it out of self preservation but I just want to. Caddy is a huge pipeliner. What most servers do is, they close the connection the moment the second request is sent / marking the second request invalid. Clients support that behavior and will do the second call under the hood.

My new proxy server has exactly the same dashboard as the previous one but the many statistics that it has have zero performance impact. It persists every second a quick snapshot to database, every 60 seconds a big one. The statistics are persistent for when the server reboots.

What an adventure.

I've had my report invaded recently.

One of the other testers made the choice to show me great disrespect by editing it, because he found a mistake in it.
For one, the fucker should've told _me_ that I have a typo, instead of correcting it himself.

It's !a fucking collaboration. I have been the one who found && documented the issue. Provided all the necessary details. Had done all the work. But this fucker makes changes in the report as if trying to portray it as a collaboration. 'Look at me, I'm being helpful'. No, fucker. You've entered _my_ domain. You have no claims here.
_My_ report, _my_ responsibility.

Now, you _might_ think that a correction was still necessary. Well, sure, I'm !claiming that I'm perfect. On a _rare_ occasion, I do have a tiny mistake.

Here's the second aspect of this.
The typo was in the 'What you should see once you've followed the Steps To Reproduce section to the letter'.
There were three outcomes: one major, && then two other of secondary importance.

The mistake was in one of those two secondary ones where I wrote 'A' instead of 'B'.

The best thing about that was that the secondary result isn't even that important.

The thing on which people should be focused is the _root cause_ of the issue. The thing occurring just before the issue is encountered. Once that thing is fixed, it is of no significance whether my mistake was there - that behavior is still gone as part of the fix.

So yeah... working w/ some credit grabbing fucks.

JoyRant build 50.

I don’t know who else is still using JoyRant but according to TestFlight there are currently 73 users (wtf?!).

The new version doesn’t ignore the cert anymore because it’s been fixed.

And the new highlight is: auto-retry!
If a Slim Shady Error occurs, the app will retry up to 5 times until giving up and showing the error.

It’s really smooth again now. What a difference!

The retry feature is implemented on the SwiftDevRant SDK and is currently set to 5 by default.

@retoor if you want to use it, just update to the latest version 2.3.0
It’s better than just hiding the error because as a user you don’t need to manually repeat the request.

Have fun!

Have I ever told you how I found this hilarious && stupid debug feature in one of my old $company's released games?

It was during the Android 4.3 Jelly Bean era. I had very little to do w/ the actual $project_0, but was tasked to test it, briefly, so I was aware of its existence.

Later on, when I was testing $project_1, I would test the chat _a lot_. So much, in fact, that I purchased a keyboard that I would connect to the Android devices to help w/ the whole testing.

Just for fun, I downloaded the $project_0 from the Google Play Store && started messing w/ it using the newly purchased keyboard.

...turned out that a debug feature had been unintentionally left in the game. During the tutorial, you could press '0' on the keyboard to skip it...

Nothing major, to be sure, but still, fucking hilarious. Given how the $project_0 had already been left to die a slow death && was no longer in development, I mentioned the issue to the PM as a curiosity.

Too bad the original programmer was no longer working for the $company.

I'm sure he'd have a 'WTF?' moment.

Next display server protocol proposal that would succeed Wayland has been just announced:

Every GUI elements require 3D graphics API and live ray tracing because it looks "cool" and "awesome".
Screen buffer won't get pushed to the display unless it's been perfectly ray traced, not missing every spot.

Users are required to enter their password everytime an application calls the drawing API to prevent unauthorized memory modification.

The mouse cursor is delayed about 10 seconds because the floaty feeling is the "future". It will also follow your hand movement with a 500ms interpolation curve to enhance elegantness.

How did we end up with such a bad convention for font weights?

Is thin thinner than light? Is regular thinner than medium? Is medium or regular the default? Is semibold thicker than medium? Is heavy thicker than bold? What about black? Why is it even named like a color? What the fuck is going on? Just use freaking numbers!

0 should be the default.
Negative numbers should mean thinner and positive numbers should mean thicker.

Why I still don't fully believe in vibe coding: Even the frontier models are constantly wrong!

Claude says c# structs with LayoutKind.Auto can have differing layout for each usage or JIT compilation. ChatGPT says they can't (and that's what I assume is true)

Claude says to "Multiply by a large prime", 0x9e3779b97f4a7c15
...That's NOT a prime!

AI generated code looks really convincing but they always make these little mistakes. And even if you inspect the AI code yourself (which let's be honest, we're all too lazy for that :P), will you be able to spot these? I really doubt it

A few days ago: ahh christmas vacation. Let's see what we can do with the home network.
Kubernetes 1.31, it's a bit old, let's upgrade to 1.35.
Yeah good luck. Probably my custom configuration messed up the upgrade. Cluster not reachable. But glory to the brave! Let's move from bare metal to proxmox.
Ok. Ceph starts quite easy and I seem to have a general plan how to setup the new virtual nodes on my now proxmox machines.
BUT! The long vacation period is nearing it's end and I will not reach the deadline that I actually care about.

I know from experience that configure my home network at evenings is not so fun when working on deployments at work.
I know I had a thread somewhere but it seems to be lost. Anyway...

Welcome back to work everyone.

I am livid and humans are just vile and disgusting beings

you know how much we say cortisol is stress and bad for you

it isn't. cortisol literally makes you unable to feel fear, it literally heals you. it STRICTLY does this. it is NOT the "stress" response

if peoples level of cortisol is high that's GOOD, not bad. they literally have more self control and their blood sugar levels are better and their brain is protected from the ACTUAL stress chemicals. your body releases cortisol to RECOVER from stress and to ensure you can remain sane and functional during high stress, like a well-trained combat soldier and shit

I just can't. humans are so disgusting. the cult of science is just vile. they just parrot each other and perform fucking rituals blindly, calling themselves high status and pretending they have the answers to everything when they're literally backwards. never do they analyze anything they say. they ignore all exceptions to their dogma. worse than any religion in the history of mankind

Wauw, notebooklm is such an amazing app. First you had to do research yourself to deliver things for the podcast you're about to generate. Now it'll fetch just ten sources for you when you pick a subject and generates the podcast based on that. I just generated a podcast for 30 minutes about the history of computer science. It was amazing, many things fell into place for me. That 30 minutes felt like five. So amazing how the dialogue goes from the podcast people, such human interactions. A joy to listen. I prompted to make a bit story out of it. It's a great way for low effort learning; just like AI in general. Learn stuff you normally wouldn't take the effort for.

We live in magic times. My favorite at products:
1. Perplexity
2. NotebookLM
3. Claude Code
4. Suno (Made many funny songs for and about friends and family members what was appreciated, also, don't forget the song that dissed every active devRant member).

Claude code is sick, but if we just had perplexity we still had a way to learn things fast but just had to type it ourselves. Well, never hated that. NotebookLM on two because I'll learn a subject every night, maybe something not related to programming. That that shit is free, damn. Are those people not just terrible at marketing? Fine if we're hooked on AI on the future, by then it's probably cheap as fuck.

Thanks for listening to my happy rant.

HOW FUNDS RETRIEVER ENGINEER SAVED VICTIMS FROM A CYBER ATTACK
I noticed it on a quiet afternoon files missing, passwords rejected, strange alerts popping up on my screen. My heart sank. Someone had broken into my digital life and taken more than money; they had taken my data photos, documents, and records I couldn’t replace.I didn’t know where to turn until I was connected with FUNDS RETRIEVER ENGINEER Instead of panicking or promising miracles, the engineer listened. They explained that recovery wasn’t about fighting attackers, but about understanding what happened and acting quickly and carefully. They secured my accounts first, locking down access so the damage wouldn’t spread. Next came the investigation. The engineer traced unusual activity, identified when my data was accessed, and worked with service providers to flag and isolate compromised systems. I remember the relief of hearing, “We’ve found the trail. Piece by piece, my data came back. Backups were restored. Stolen access keys were invalidated. Accounts were verified and reopened. What felt lost forever slowly returned. But the most important part wasn’t just getting my data back it was learning how to protect it. The engineer helped me strengthen my security, recognize warning signs, and rebuild confidence online.What began as a cyber attack ended as a lesson in resilience. Thanks to FUNDS RETRIEVER ENGINEER, I didn’t just recover my stolen data I regained peace of mind.Contact info:WhatsApp / Telegram: +1 (802) 952-3470 

I got my House loan of $750,000.00 USD from Mark Davson Financial Loan Firm +447300935913

I am out here to spread this good news to the entire world on how Mark Davson Financial Loan Firm and his teams help me get a loan of $750,000.00 USD.
Mark Davson Financial Loan Firm has made the process of applying for a loan easy, and seamless, and it was so pleasant to work with Mark Davson Financial Loan Firm.
Mark Davson has also been great, attentive, and so helpful in applying for a Loan, and finally Mark Davson Financial Loan Firm has made it possible for me to get a House loan of $750,000.00 USD at the rate I was hoping for and he has truly gone above and beyond in helping with this process, making it easy and making me feel looked after.
Working with Mark Davson Financial Loan Firm has been such a seamless experience from the very beginning! Wouldn't have been able to get the house I wished for without the great service and support of Mark Davson Financial Loan Firm, Couldn't recommend Mark Davson Financial Loan Firm enough! Thank you Mark Davson!

I have already recommended them to others, as we are very happy with them.
If you are looking for any kind of loan, i advised you to contact Mark Davson Financial Loan Firm with the below emails:

MARK DAVSON FINANCIAL LOAN FIRM

WhatsApp:+447300935913

When you first sign up for any betting platform, the login process can seem daunting if you’re not familiar with it, and a great resource that explains how to access your account on Pin Up Sports Betting is https://pinup-sports-betting-bd.com/... which provides step‑by‑step instructions on signing in, recovering forgotten passwords, and ensuring your account remains secure. Knowing exactly how to log in helps you start placing bets faster and reduces common hurdles that new users face.

Logging in securely starts with using the correct credentials and ensuring your browser or app is up to date. Many people don’t realize that outdated software can lead to login errors or issues with displaying the site properly. Using a direct guide to walk you through the process eliminates guesswork and confusion, especially on your first attempt.

Another aspect covered in reliable login instructions is account security. Options like two‑factor authentication, email verification, and secure password creation are often recommended to protect your account from unauthorized access. While it might seem like an extra step, taking advantage of available security features ensures peace of mind when playing or wagering online.

Finally, once you’ve successfully logged in, understanding the dashboard or main user area is crucial. Knowing where to find your account balance, how to navigate to various betting markets or games, and where to access support or promotions makes your experience smoother and more intuitive. Getting comfortable with the login and interface frees you to focus on strategy and enjoyment rather than technical frustration.