I’ve been freelancing on the side for more than couple years now, mostly helping startups and smaller teams fix bugs, add features, the usual stuff.

Used to be maybe 1 or 2 projects a month. Now I’m turning people away because there’s too much work coming in. And I’m pretty sure I know why.

About 70% of the requests I get now are basically “we built this with AI and it doesn’t work, can you fix it?”

tbh I’m not mad about it. The money’s good and the issues are usually pretty straightforward once you dig in. Last few weeks alone I’ve seen zero input validation, hallucinated libraries that don’t exist, payment logic that does the opposite of what the comments say. The security stuff is wild. Apparently 45% of AI-generated code has vulnerabilities and I believe it.

Don’t get me wrong, people hired me to clean up messy code before AI too. But it used to be like 1 in 10 projects. Now it’s most of them. And the …

At work I have non-technical business managers dictating what softwares to make. And these aren’t easy asks at all — I am talking about software that would take a team of engineers months if not an entire year+ to build, but as a sole developer am asked to build it. The idea is always the same “it should be simple to build”. These people have no concept of technology or the limitations or what it actually takes to build this stuff — everything is treated as a simple deliverable.

Especially now with AI, everyone thinks things can just be tossed into the magical black box and have it spit out a production grade app ready for the public. Not to mention they gloss over all the other technical details that go into development like hosting, scaling, testing, security, concurrency, and a zillion other things that go into building production grade software.

Some of this is asked by the internal staff to build these internal projects by myself and at unrealistic deadlines - some are just flat …

I experienced a debilitating case of burnout a few years ago and never fully recovered. After a lot of reflection, I’ve realized this was partly due to the lack of mentorship I received as a junior, which immediately put me on a path of anxiety and overworking to prove myself. This just compounded over the years as I progressed and gained more responsibilities.

This industry seems to be unique in that kids straight out of college are seen as subject matter experts and immediately pressured to contribute. In my first two jobs, there were major reorgs right after i onboarded and I was immediately thrown into the fire. I had to navigate the workplace environment and culture by myself, never feeling like I belonged.

In my many years as an IC, I’ve never had someone sit down with me to discuss career goals or professional development. I grew up in a blue collar environment with no exposure to people in professional fields as a kid, so this lack of mentorship affected me particularly hard… …

In my organization, the engineering managers function more like secretaries. They book meetings, handle logistical issues, conduct initial interviews with new hires (nothing technical), and set our salaries. The managers I’ve had have never had a good idea of what I’m actually doing. They don’t know much about the product and have probably never looked at the code. In my experience, whenever they try to make decisions on their own, things tend to go awry.

Is this common? I feel like it would be much better to actively encourage engineers within the teams to take on managerial roles while still doing some of the team’s actual work. But about 4 out of 5 manager hires in my company come from outside. Maybe it’s a Sweden thing.

Recently, I gave a talk and a lesson at my old secondary school to a class of 14–16 year-olds learning IT - simple things, learning HTML, and some CSS.

When I asked how many wanted to be software developers, almost all of them raised their hands. I didn’t have the heart to tell them the job market’s cooked and most of them probably won’t find work easily.

One kid asked if I regret becoming a developer. I told him no, that the money’s great, especially as an owner, but I couldn’t really answer the question.

It’s just gotten kinda boring. Everything’s “AI this, AI that.” Before that, it was “big data,” before that something else. It just feels like we’re constantly chasing the next buzzword. Cyber-Sec is probably the next big thing. Now, the hotshit is in data-analytics.

I honestly had more fun showing them how to build bar-charts in Jupyter Labs using Python and teaching them simple SQL syntax than I did in the last couple of weeks attending client meetings. It just brought …

The tablet itself has only 1GB RAM but I still managed to make do by allocating 512MB RAM on a Paper 1.8.8 server.

It’s been a while since I last posted here, but I’ve got something cool to share. This is a fully self-hostable, open source overlay network that comes with a slick visualization tool for your remote access policies.

Basically, you can spin up your own overlay network to connect your homelab or org resources, and then actually see how access is structured with multiple views:

Peer View → see what groups a peer can access + which policies allow it

Group View → check which groups/users can access resources

Networks View → explore which peers/groups can access specific networks/resources

Go check it out on GitHub: https://github.com/netbirdio/netbird?tab=readme-ov-file#quickstart-with-self-hosted-netbird

I thought running my own setup would be cool and save me time, but now I’m stuck dealing with logs, weird configs, and constant updates. Does anyone actually get to enjoy their server, or is everyone just fixing stuff ²⁴⁄₇ like me..

So I started self-hosting immich, and it was all pretty good.

Then today I wanted to download an album to send the photos to someone - and I couldn’t. Looked it up, and it’s apparently the result of an architectural decision to download the whole album to RAM first, which blows up with anything over a few hundred megabytes. The bug for this has been open since December last year.

There’s also the issue of stuff in shared albums not interacting with the rest of immich - searching, facial recognition, etc - because it isn’t in your library, and there’s no convenient way of adding it to your library (have to manually download/reupload each image individually). There’s a ticket open for this too, which has been open several years.

This has sort of taken the shine of immich for me.

Have people who rec it here overcome this issues, never encountered them, or don’t consider them important?

Bait title is bait.

But essentially, I want ad breaks for my kids/me to keep ontop of things and have healthy breaks of activity during the days we lounge about.

Ideally these would be mandatory/unskippable “ads” that play for 30s-5min and either display a living task list from HA of chores they need to do or (for me) a brief workout challenge like “wall sit for 3 minutes / do X amount of push ups” via a static image or video file.

Is there any way to accomplish this?

Hi all, casual Usenet user for many years now. Just recently in the magazines and comics groups I’ve been seeing a ton of files coming over with a .cne extension and the word “IRap” afterwards and the letter XO or ZO. Is that someone spamming the group, or is it intended to be an actual file? Most of the files are incomplete as well. Any info on this?

I am trying to sign up for NZBcave, and it’s telling me I need to get a confirmation code in my email before I can finalize… but it tells me before it will do that I have to read a post by the administrator… which I can’t do until I get the confirmation code… and around and around and around I go.

Anyone else have this issue? The Help link on the page just goes to the VBulletin FAQ, no way to reach out :/

I am a long-time internet user since the dialup BBS days on my Commodore 64. Decades ago I spent some time on usenet but never really found it to be my thing. I’ve recently retired and am trying to keep my brain active by learning new stuff. I’ve done a lot of reading here have what I think is a decent grasp of how backbones, providers, resellers and companies work, but I have some simple questions about file transfers. I hope you’ll indulge an old man who’s trying to get caught up.

1. I understand that file transfer involves converting binaries to text and then reassembling them into binaries after download. Is this accurate?
2. Various usenet resellers set different price structures for downloading. Say I have an account that has a 40gb monthly download limit and I wish to download a large binary file, say an ISO of a custom Linux distro that is 100GB. How does the conversion/reassembly work? Will I end up downloading 100GB of text and thus exceed my monthly …

• 2TB Prime Block + 2TB XS Block | $32
  • 2TB + 2TB (deal link here)
• Prime Yearly Unlimited + 1TB XS Block | $30
  • Prime Yearly Unlimited + 1TB XS Block (deal link here)

We’re kicking off another Prime Day with exclusive Usenet deals just for our valued members! Whether you’re a seasoned user or just getting started, now’s the perfect time to upgrade your plan, lock in savings, or explore new features.

🔥 Limited-Time Offers Include:

• Deep discounts on high-speed access
• Bonus data packages
• Special pricing on extended plans

📅 Hurry — deals end soon!

🙏 Thank You, Members!
We appreciate your continued support and loyalty. Prime Day is our way of giving back — enjoy the savings!

For those of you who don’t know, I’ve been working on an app for Plex users who have a ton of content and don’t always know what they want to watch. It’s super-easy to use, requires no additional setup or configuration, and it’s really fast.

It’s called Coax (like the cable), and I just got it working on AppleTV. The beta is already running on iPhones, iPads, and Macs (native Vision Pro coming sometime, probably?). I’ve been working hard on it and I’d love to hear what you think of it.

Here’s a Google doc where I’m capturing my plans for V1 feature development. I’m also using it as a bug tracker, along with the TestFlight feedback. If you want to join the TestFlight beta, you can do so here: https://testflight.apple.com/join/fWRGNxPu

A couple of things I want to make sure folks know before they sign up:

• This app is still very much in development, so please send in feedback. Feel free to DM me, comment on the doc, or …

Saw this while going into the Playback Settings today

So we know that Plex can download media in advance you prevent buffering, but if you need to jump back say 30 seconds, it has to reload again. Why can’t it keep the last minute or two in memory so if you jump back 30 seconds then it’s instant? That combined with 5 minutes in advance would make having to change seek point more fluid. Thoughts?

I used plexam for years. Switched to spotify for 3 months. Then apple music for 6 months. Now i am using both plexamp and AM

IF your library has more than say 10k tracks, nothing comes close to plexamp. Playing a fav album with dj stretch is simply unmatched experience

Hi All, I got sick of doing this manually and 99% of what I need from TDARR was just to reduce file sizes and keep quality. I had this as a bash script and decided rewrite it in golang.

It interrogates the existing file and matches the quality or just slightly better.
Keeps all Audio and Subtitle tracks as well as chapters etc.

It’s already transcoded about 17TB of media into less than 7TB for me.

Supports hardware encoding with FFMPEG and can basically be built for any architecture.

I’ve supplied an AMD/x86_64 Binary in the bin directory for the 90% of you out there running that hardware. (ie just copy that file, chmod +x it and you can run it)

Pro-tip, use an SSD backed working directory and hardware encoding and you can max out your local IO or any ¹⁄₂.5/10Gbit link to your media box if you have one.

Hopefully helps somebody.

https://github.com/lancestirling/htoh