Dax Raad from anoma.ly might be the only CEO speaking honestly about AI right now. His most recent take:

“everyone’s talking about their teams like they were at the peak of efficiency and bottlenecked by ability to produce code

here’s what things actually look like

- your org rarely has good ideas. ideas being expensive to implement was actually helping

- majority of workers have no reason to be super motivated, they want to do their 9-5 and get back to their life

- they’re not using AI to be 10x more effective they’re using it to churn out their tasks with less energy spend

- the 2 people on your team that actually tried are now flattened by the slop code everyone is producing, they will quit soon

- even when you produce work faster you’re still bottlenecked by bureaucracy and the dozen other realities of shipping something real

- your CFO is like what do you mean each engineer now costs $2000 extra per month in LLM bills”

Im mid-level at a standard non-tech Fortune 500 and the overall mood seems mildly checked out. Most devs are offloading a lot of their work onto Claude. It’s not slop. It’s reviewed, refined, and tested, but it is still reducing intimacy and familiarity with the repos.

People are mostly camera off. A lot of people are ignoring the in office mandates. I’ve noticed more gaps in slack response times which leads me to belief people are off doing things during work hours (and to be clear, I’m fully fine with this. In an ideal world that is the what AI is supposed to enable).

Regardless, the work is getting done, the stock is doing well, the company is in good shape financially. But the general mood and enthusiasm is just mildly resigned, at least on the Dev side.

Wondering if this is common.

I’m seeing a trend where many devs stop using their brains and just trust Claude to do their job. They generate the code, push the PR, and let reviewers review “their” work.

As a reviewer, I’m honestly tired of pointing out bullshit 500+ lines super verbose Claude slop code filled with complicated logic and small mistakes that could’ve been avoided if the author wasn’t too lazy to review their own work properly.

Think about it. These people are basically offloading their work onto the reviewers. All they did was write a prompt in basic English and pass the generated code along. Then reviewers do the hard work of actually reviewing and fixing what they were supposed to write in the first place. After it’s merged, they still get to claim the full credit. Reviewers get nothing.

To put it simply, they’re just middlemen between Claude and the reviewers. So why not just assign the task to me, let me prompt Claude directly, and let me claim the credit instead?

Maybe …

Am I completely missing something?

I use LLMs daily to some context. They’re generally helpful with generating CLI commands for tools I’m not familiar with, small SQL queries, or code snippets for languages I’m less familiar with. I’ve even found them to be pretty helpful with generating simpler one file scripts (pulling data from S3, decoding, doing some basic filtering, etc) that have been pretty helpful and maybe saved 2-3 hours of time for a single use case. Even when generating basic web front ends, it’s pretty decent for handling inputs, adding some basic functionality, and doing some output formatting. Basic stuff that maybe saves me a day for generating a really small and basic internal tool that won’t be further worked on.

But agentic work for anything complicated? Unless it’s an incredibly small and well focused prompt, I don’t see it working that well. Even then, it’s normally faster to just make the change myself.

For design documents it’s helpful with catching …

I was set with a technical interview titled with “code review” and no context.

I was really looking forward to this company as culture was chill and pay was lucrative. Remote too.

On call, both interviewers were very cold and presented me with a frontend feature and then said, how would you navigate this feature on a new code base.

Basically I just had to guide him like a junior developer on his screenshare.

And, my personal way of development is very CTLR + F heavy. I just asked him to random things and I felt like I utterly bombed the interview. In real life, I navigated lots of code bases but in this particular interview I just forgot how to do it.

I feel so stupid like 7 YOE and can’t even do code navigation on new project.

They ended interview 25 mins earlier than scheduled time and very abruptly brought in “do you have any questions”.

I was awestruck and I couldn’t even ask any questions. It was so embarrassing that it hurts.

Seerr is the new unified successor to Overseerr + Jellyseerr. The two teams have merged into one project + one shared codebase, combining all existing Overseerr functionality with the latest Jellyseerr features, including Jellyfin + Emby support.

Highlights

• Jellyfin + Emby support (alongside Plex)
• Optional PostgreSQL support (in addition to SQLite)
• Blocklist (movies/series/tags) + Override rules for smarter request defaults
• TVDB metadata support (experimental) + TVDB indexer
• DNS caching (experimental) to reduce DNS spam (Pi-hole/AdGuard friendly)
• Dynamic placeholders in webhook URLs
• Notification QOL (e.g., optional embedded posters) + lots of bug fixes

Migrating from Overseerr/Jellyseerr

You must follow the migration guide linked below carefully. BACKUP FIRST so you can roll back if needed Release notes: https://github.com/seerr-team/seerr/releases/tag/v3.0.0

Release announcement: https://docs.seerr.dev/blog/seerr-release
Migration guide: https://docs.seerr.dev/migration-guide …

So 2 weeks ago i posted about my risky move where I bought two dying hdd from Vinted that were still under warranty and sent them to seagate for replacement.

579 people votes and almost 50% thought I wouldn’t get a replacement.

I’m happy to say that seagate has sent two replacement HDDs in perfect Health 😎

SAAS. The Warner Brothers acquisition. Ads. I’m so tired of it all.

Now it’s been a month and a half since i started work on this humble home server.

It currently consists of:

… an HP EliteDesk 800 G3

• CPU: i5 7500 3.8 GHz
• RAM: 16 GB DDR4
• SSD: 256 GB M.2 + 4 TB 2.5”

… running Arch Linux

• yes

… hosting a Jellyfin stack

• for my Linux ISOs

… inside Docker containers

… which I, gf and family connect to through Tailscale

Edit: The Arch Linux pain is brutally overexaggerated in my limited experience. Do correct me if you’ve ever had a basic Jellyfin/Docker setup break on an update.

I vibed-engineered this to solve a problem at home and I’m sharing it in case other families here can use it. First open source project, so feedback is welcome.

The problem:

I wanted my kid to use YouTube for learning, but not get swallowed by the algorithm. Every existing solution was either “block YouTube entirely” or “let YouTube Kids recommend whatever it wants.” I needed something in between — a gate where I approve every video before it plays.

What it does:

Kid searches YouTube via a web UI on their tablet → I get a Telegram notification with thumbnail, title, channel, duration → I tap Approve or Deny → approved videos play via youtube-nocookie.com embeds. Pair it with DNS blocking (AdGuard/Pi-hole) on youtube.com and the kid can only watch what you’ve approved.

Stack:

• Python / FastAPI + Jinja2 (web UI)
• yt-dlp for search and metadata (no YouTube API key needed)
• Telegram Bot API for parent notifications + inline approve/deny buttons …

Edit: Github page updated

https://github.com/Hemeka/Discord-Alternatives

I’m curious how providers will keep their retention and old posts when hard disk drives are so expensive. I’m worried posts are going to go bye-bye. Any Usenet providers care to chime in?

I’m an old guy which my computer journey started with BBS/usenet newsgroup/VMS notes. So, supposedly I knew what is usenet news. In that good old days, usenet news were free, usually, local ISP or university computer center will have a server, then, use Thunderbird (or other text based program, like Euroda) to read news in various groups. Come straight to my question #1.

Is this kind of free to access newsgroup server vanished? Only subscription based services remain?

Then, contents, I participated in a lot of newsgroups in the past, mostly text discussion in whatever topics. I admitted that I read abpe, but need to assembly millions (I know millions is exaggerated!) of uuencode files into one GIF file. I knew how to combine all those uuencode files into one and run uudecode on it. My question #2.

I read in Rules/FAQ that binaries/movies/copyrighted materials should not be discussed, does it mean those items are now carried in usenet newsgroup? Then, are there zillions …

title

Hi

Im looking for help to maximize my speed.

I have Newshosting, Newsdemon and Eweka but I rarely get an above 1G speed when downloading.

I have cat 8 cables.

2.5 gig tp link switch.

Ex280v tp link router.

Ive configured Sab and all downloads to a NvME drive.

Any help or suggestions and how to get a higher than 1Gps would be great,

Thanks for any help provided.

Been a long time user of Usenet but never really have understood the method of posting where a indexer like Dog or Geek can pick up the post. Is it all in the header info that you leave in the nzb with software like ngPost? And if it is in the header then what info do you need in there and how do you get it all in there?

Like many others I had dropped frames occur when watching 4K movies on my Apple TV 2022 4K, I didnt want to switch to Infuse and figured I could probably solve it with some parts I had laying around and my 3D printer.

I wrote a blog post going in detail about how I measured the temperature and validated my solution. I included some additional details of the hardware used and how I automated it with a smart plug and Tautulli.

TLDR is just attach a fan and small heatsink (possibly optional) to the top of the Apple TV and it keeps it cool off to prevent frames from dropping.

I used a 120mm pc fan and a 10mm thick cob led heatsink I had laying around. Then I designed and printed a rack mount that houses it all in a neat package. I’ve included a model that isnt rack mounted if you have your Apple TV sitting on a shelf.

Hope this is able to help anyone who has similar issues.

Hi there Reddits

ChuckPA is signing out!

Should the name ChuckPA ring a bell to you, then scroll down

And if not, then read ahead:

* ChuckPA is a Linux Expert

* ChuckPA became a Plex Ninja

* ChuckPA became a Plex Team member (Contractor)

* Due to his skills, ChuckPA wrote almost every script for installing Plex on a Linux platform, including NAS boxes

* ChuckPA handled implementation of GPU’s on almost every Linux platform via scripting

* ChuckPA created multiple opensource scripts to help you with Plex, like the DBRepair and the ResetCredentials scripts

So if you think that Chuck helped you, then please do not respond here, but instead tell him here: https://forums.plex.tv/t/eosl-notice-chuckpa-transition-to-legacy-support-mode/936513

Thanks

I built an app that turns your Plex library into a 90s-style cable TV experience

NostalgiaTV is an Android TV and mobile app that turns your Plex library into a fully simulated retro cable TV lineup.

Instead of choosing what to watch… you just tune in.

It automatically builds channels, schedules shows, and gives you a classic 90s-style program guide so it feels like flipping through real cable again.

For the past couple weeks, nearly 100 testers have participated in the beta testing phase and have given me tons of feedback and feature requests. Nearly all of them have been incorporated into the app to make it feature-packed. I’ll be resetting your app access now. This will not remove all your custom channels, but it will take you off of the tester list. Again, thank you for all your support! :)

If you want to check it out, it’s on the Google Play Store available for download here! You can either try out the demo or try the first 10 channels for free!

What it does …