Search: Cloud/dev tooling

I replaced cloud LLMs with local models running off a Proxmox LXC, and the performance trade-off was worth it

… On my aged system, I simply ran the ls -l /dev/nvidia command to get the device IDs 195, 235, and 237 for my GPU , pasted the following syntax into the LXC’s config file, and installed the graphics card drivers inside the LXC to configure GPU passthrough, before compiling llama.cpp’s Vulkan variant… …

May 29, 2026 · Ayush Pande

I ran Gemma 4 (26B) on a 10-year-old-GPU, and it's reliable enough to replace the cloud

… I’d already done this for my host machine with the following commands, so all I had to do was pass them to the freshly-configured LXC. wget https://us.download.nvidia.com/XFree86/Linux-x86 64/580.119.02/NVIDIA-Linux-x86 64-580.119.02.run chmod +x NVIDIA-Linux-x86 64-580.119.02.run ./NVIDIA-Linux-x8… …

May 10, 2026 · Ayush Pande

I stopped paying Heroku for my app deployments after discovering this self-hosted alternative

… Related 4 reasons you should create a dev VM if you’re even remotely into coding Virtual machines are a game-changer for coding enthusiasts But it’s just as effective for Docker-based projects You can even set up conventional FOSS utilities on Coolify Coolify may be more dev-centric than most conta… …

Apr 19, 2026 · Ayush Pande

I replaced Cursor and Antigravity with a completely local VS Code setup, and I missed less than I expected

… Meanwhile, I’ve started experimenting with MoE models, and with the right extension on VS Code, they’ve entirely replaced their cloud counterparts for my dev tasks. …

Jun 3, 2026 · Ayush Pande

GPU passthrough to LXCs beats VMs in Proxmox, and it's way simpler than you'd think

… For my Nvidia-powered rig, I tend to add the following lines at the bottom of the file, but you might want to change the number after the lxc.cgroup2.devices.allow: c argument to match the GPU components from the output of the ls -l /dev/nvidia command. lxc.cgroup2.devices.allow: c 195: rwm lxc.cgr… …

Mar 26, 2026 · Ayush Pande

Self-hosting freed me from the OS lock-in trap (and made my devices clutter-free)

… Heck, you’ve got handy tools like AirDrop that remain inaccessible outside of Apple’s premium devices. Self-hosted applications on a local server spare me from the OS problem, without the privacy-intrusive aspects of their cloud-based counterparts. …

Mar 28, 2026 · Ayush Pande

You need these 6 self-hosted tools if you value your privacy

… SearXNG See at Github Home Assistant A central hub for your smart home Despite making your life more convenient, smart devices that connect to the cloud are riddled with privacy and security problems. …

Mar 26, 2026 · Ayush Pande

This 18-year-old's free Lightroom replacement blows Darktable's UX out of the water

… Its absence in RapidRAW is a commendable feat for experienced software developers, and more so, for one teen. Yes, the community is steering and assisting in the development of this well-documented project, but it's already usable on Windows, macOS, and Linux, with Android builds in the pipeline. …

May 1, 2026 · Chandraveer Mathur

I stopped choosing between VS Code, Google Antigravity, and Cursor and built a hybrid workflow that uses the best of them

… If I’m working on something that requires spinning up heavy service stacks, database containers, or intensive cloud-aligned architectures, I don’t want my laptop fans spinning at maximum speed. Running it on Antigravity’s cloud workspace infrastructure leaves my local machine unburdened. …

May 20, 2026 · Parth Shah

Instead of Claude and Anthropic models, I use my local LLMs for coding (and not vibe-coding, mind you)

… As a home labber who loves tinkering with DevOps projects, I conduct most of my experiments on my local nodes, and I’d rather not let some cloud-based LLM gain access to the credential files, API tokens, and server configs I keep in my Terraform and Ansible project directories. …

Mar 23, 2026 · Ayush Pande

Followed topics