For a while, one small thing was driving me crazy every time I powered on my Proxmox machine. Not the CPU. Not the storage. Not the software. The fan. It was loud. Constantly loud. Full speed from the moment the machine started. And once I noticed it, I couldn’t ignore it anymore. That’s the annoying part about hardware problems like this — they don’t stay in the background. They slowly take over your attention until fixing them becomes the only thing you can think about. So I started digging. I checked the socket type. I removed the motherboard. I measured the mounting holes. I confirmed what I needed. Only then did the search begin. The setup itself was nothing exotic: an Intel Core i5-6500 sitting on a Lenovo ThinkCentre M900 motherboard, model IQ1X0MS (30BC), with an LGA1151 socket. A budget CPU on a budget OEM board — simple on paper, but not so simple when the fan refuses to behave. At that point, the goal was clear. I didn’t need the biggest cooler. I needed the quietest practical solution. So I compared a few options: Noctua NH-D15 - extremely powerful, but probably more cooler than this CPU really needed. be quiet! Dark Rock Pro 5 - premium, silent, and also a bit overkill here. Endorfy Fortis 5 - a strong, balanced option with good value. Endorfy Fera 5 - compact, affordable, and practical. be quiet! Pure Rock 3 - quiet, sensible, and enough for a 65 W processor. And after thinking it through, the choice became obvious. I went with the be quiet! Pure Rock 3. It was reasonably priced was more than enough for the i5-6500.And it solved the problem without turning the build into something unnecessarily expensive or oversized. But the cooler alone was not the whole story. No more background irritation. No more fighting the machine every time I wanted to use it. Just a simple fix that made the whole setup feel right again. And honestly, sometimes that’s the best kind of upgrade - not the flashy one, but the one that quietly removes a problem you were tired of living with.

I recently switched my Omarchy Linux setup to give QEMU/KVM VMs direct LAN access using a native bridge (br0) and systemd-networkd. Here is the exact config I used to get it running without sudo. I dropped three files into /etc/systemd/network/ to define the bridge and attach my interface (eno1). - /etc/systemd/network/10-br0.netdev [NetDev] Name=br0 Kind=bridge - 20-eno1.network [Match] Name=eno1 [Network] Bridge=br0 - /etc/systemd/network/20-br0.network [Match] Name=eno1 [Network] Bridge=br0 - /etc/systemd/network/20-br0.network [Match] Name=br0 [Network] DHCP=yes Then I just ran sudo systemctl restart systemd-networkd.service To let my VMs use this bridge without root, I updated two config files. - /etc/qemu/bridge.conf allow br0 - /etc/libvirt/qemu-bridge-helper.conf allow br0 After a sudo systemctl restart libvirtd, my VMs connect directly to the network via br0 instantly. Clean, native, and no sudo needed for launch.

Hello devs, Welcome to this communiy and my post. I am a senior Full Stack and AI Engineer, currently I am gonna post my previous works. There are quite a few full-stack, AI, and software engineers in this community. I would be very grateful if you could leave comments sharing screenshots, GitHub links, or links to project websites of the projects you have previously developed. This will be beneficial for mutual cooperation, fostering opportunities, securing jobs, and the development of the community. I am presenting my work first.

I started building an Obsidian plugin and wrote almost every line of it with AI, using DeepSeek v4 inside OpenCode. The speed is real. Development flies, and it delivers features really, really fast. But the key word is almost. I still have to review, fix, and adjust the code. The problem is that AI does not care about design patterns, naming conventions, or consistency. It produces code that is hard to read. Code nobody cares about is bug-prone. When it breaks, I am the one who has to fix it. So, I made the tooling care for me. Every check with a strict mode runs in strict mode: TypeScript: A silent type bug becomes a failed build seconds after I write it. It never ships. - Prettier: The formatting debate that used to take a team meeting now takes one command. - Knip: Dead code is flagged the moment it is no longer imported. I delete it and move on. - jscpd: Paste a block twice, and it tells me to extract a shared function. - depcruise: One architecture-breaking import kills the build. My design survives. The real hero is AGENTS.md. I wrote down exactly what I want the AI to follow during development, focusing primarily on code structure. The model reads it before writing a single line. One file replaced a hundred code reviews. The payoff is immediate. I can check the code anytime and understand it in minutes, not evenings. When the checks pass, the code is mine, even though I did not type a word. The lesson is simple. Quality gates are not overhead. They are what makes AI-written code yours. The AI brings the speed. The checks bring the judgment. You bring the standards. Here is the link to the repository so you can check out what I did: https://github.com/Karol-Szykula/NotePalace/tree/develop