Ive been seeing obvious bot activity, weird upvote/downvote activity, and overall just a weird vibe from here. I honestly think half the people in this sub and similar subs arent real people. Pretty depressing to think about and makes me want to just delete the whole app. Am I being paranoid or are we firmly in the dead internet right now?

I think we all have our a-ha moments at work when we try a new approach to something, and it just clicks, makes you love the job more, and increases your output. improves the quality of your work, etc.

Here are some examples from me:

* Learning to stop working at some point and avoid hyperfocus. This only makes it worse, energy-wise, the next day

* Documenting all decisions and important calls religiously. I don’t do anything fancy, just lots of Google Docs that I immediately share with all the parties involved

* Following advice from Dan Luu, I am ready to look stupid if something sounds off and too vague. Very often, the others don’t understand it either. Occasionally, however, that does make you look stupid

* Sort of contradictory to the previous one, not saying anything at meetings unless you have a better alternative than silence is also something I try to do. Unfortunately, some developers try to mark their presence anyhow.

* Communicating as much as reasonably …

The 2010s tech culture conjures up images of a relaxed office space with bean bag chairs, ping pong tables, and a snack bar. That whole chill Silicon Valley vibe. But now? It’s quite a stark contrast, almost polar opposite… Even before AI, the tech space has just felt like a constant anxiety trip with fears of being laid off, stacked ranking+forced attrition, expected to work nights, weekends and holidays. Everyone in tech pushing the whole GaryV + Goggins grindset. It has become increasingly toxic.

What the hell happened?

Something I’ve been noticing over the past year and I’m curious if others are feeling it too. Our team measured roughly 4-5x speed improvements on individual coding tasks with LLMs. But when we looked at total project delivery time, it was maybe 1.5-2x faster since we enabled claude code and got cursor licences.

The gap bugged me for a while so I took a gander at our project management tooling and the tl;dr is that it’s all went into doing the work that surrounds the coding. More and more do I feel that programming is shrinking as a percentage of our weeks, and what’s replacing it looks a lot like product management. Orchestration, prioritisation, communication - more of a PM role. I’ve been in this for a little while, but I’m seeing juniors ‘speedrun’ past the SWE best practices. Right now it’s clearly backfiring, but will it in a year or two?

Anyone else tracking this? I posted this on /cscareerquestions, but didn’t get …

I’m a senior SWE (remote, which I suspect may be important here) who’s been trying to make staff for a few years now. About a year ago, I pitched a major project to my team lead – major changes to an underperforming data store which had caused outages. The team lead shut me down, telling me that we didn’t have time for major changes and that we needed to focus on point fixes and firefighting for individual customers.

I eventually switched teams, and because of the ambiguous ownership structure at our company, I was able to get a version of the same project approved by my new manager, who’s more aligned with me. We agreed to make it the foundation of my case for a staff promotion. I’ve been working on the project more or less solo for about nine months now, and during that I’ve tried to solicit my former team lead’s input on my work, which will still be highly relevant to their team, and I’ve been consistently ignored. I have no …

Hello everyone,

I have been self-hosting production applications (not just personal projects, but fairly decent ones with significant traffic) for over a decade.

After my last startup (advertising marketplace) failed 2 years ago, I wanted to share my knowledge with the community (which I learned everything from) since the current resources were either too shallow, lacked real world examples or didn’t address the knowledge gaps.

The book starts with the basics and builds up to covering the full infrastructure stack, with the goal of understanding the system as a whole and eventually deploying on Kubernetes. Kubernetes is a major focus, but the content can be applied to any environment. You should probably check the Best Practices section for tips on home servers.

It is available for free at the https://selfdeployment.io including the PDF and the code blocks. Yet, you are welcome to pay what you want.

As a bonus, here is my home server rack and its guardian.

Budget case with great airflow

Microsoft has just submitted this e-mail which says your data will be used to train their AI unless you explicitly opt-out.

They supposedly explain how to do it, but conveniently “forget” to include the actual link, forcing you to navigate a maze of pages to find it. It is a cheap move and totally intentional.

To save you all the hassle, here is the direct link to opt-out: https://github.com/settings/copilot/features and search for “Allow GitHub to use my data for AI model training”.

Hello !

I’m officially introducing this tool, which I now consider stable after a month of work.

I say “officially” because I’ve already discussed it in this same subreddit, after talking about my home lab and posting a screenshot of the tool.

https://github.com/Pouzor/homelable

So, what’s it for?

Well, simply to save me (and you maybe) time by creating a nice visualization of my self-hosted infrastructure.

What’s different from draw.io? I’d say mainly the ability to scan the network and manage a “library” of discovered hardware. This lets you click -> add to the diagram. The scan listens for open ports and helps “guess” which services are running on each machine (e.g., 8123 for Home Assistant, etc.), saving time when documenting the infrastructure.

Otherwise, it mostlty visual customization options and a way to import/export the configuration if you want to document it in a YAML-like format.

I also had fun testing the integration with an MCP server (optional) to feed it raw …

Ever since yesterday I’m downloading all fake stuff with exe files. Had to disable it in sonarr/radarr. Have they been hacked or something?

https://preview.redd.it/k4m3d7kcgorg1.png?width=498&format=png&auto=webp&s=8e39c6d869647dfda8efaba6e0a9fbb0003990d0

Been downloading stuff since the 14.4 kbps days. I am stunned we can basically download 1 GB per second now. 50 connections on the Europe server with my UsenetExpress block.

EDIT; 8 gbps.

hey guys,

I’ve recently configured everything on nzbget, using newshosting and nzbgeek. I’ve found a fair few things are defaulting through to qbitorrent with less than 10 seeds an takes forever.

a few requests would fail on nzbget with ‘health is 97.9% below critical 98”, but others have said 92.0 below 92.1?

how can I improve my services to prevent this? do I need another in prowlarr as an indexer?

Apologies in advance for what is probably a really simple issue to resolve.

My NZBgeek account has been flagged twice now and on asking why it was flagged again I was told “Your account made 21 requests for various content from 5 different Amazon IP addresses.

This is normally related to using some sort of service/browser plugin.

Over the last 6 days your requests have been coming from Virgin Media ISP until 20 mins ago.”.

I’m not sure why this is happening suddenly? I used SABnzb as well.

Any help would be massively appreciated.

What they removed was a way for the community to help fill in the gaps in places where Plex was lacking. It helped (and helps) build useful tools for users to have a richer experience.

Thanks for listening, Plex, and I hope you can find a way to make it work.

For reference:
https://forums.plex.tv/t/information-related-to-security-vulnerabilities/935164/56
https://www.reddit.com/r/PleX/comments/1s3t4lw/rip_plex_auto_languages_and_other_thirdparty_tools/

UPDATE: 03/27 09:07 PST - Plex Auto Languages is working for shared users again!!!

If they’re going to remove this third-party functionality, they should implement Auto Languages in Plex. It’s so annoying that Plex doesn’t remember your language selection automatically, so you have to keep switching languages when you start an episode.

The script Plex Auto Languages was a fix for that, and now it no longer works.

In an effort to keep Sonarr secure and provide some additional information we have published two security advisories for Sonarr today. Please ensure you’re running Sonarr 4.0.17.2950 or later where both these issues are patched. Additional information is provided in the advisories. If you’re running Sonarr with `Authentication Required` set to `Disabled for Local Addresses` or Sonarr is running on Windows you should review both advisories and take appropriate action.

https://github.com/Sonarr/Sonarr/security/advisories/GHSA-h5qx-5hjf-7c9r

https://github.com/Sonarr/Sonarr/security/advisories/GHSA-h393-v5hm-6h8f

If you believe you’ve found a security issue you can find more information on reporting it to us on GitHub. https://github.com/Sonarr/Sonarr/security

I love Plex. It’s possible for me to self host anything and everything I want.

My wife loves a scroll. She loves going through Netflix or any other streaming service and just seeing what’s about.

No matter what self hosted new platform I use, nothing replaces the algorithm of these services.

Am I doomed to pay forever, as that is easier than changing my wife’s habit.