I finally stopped forcing local LLMs and switched back to cloud AI
…While newer models now have MCP server support and all of the fancy stuff, not every local LLM supports them, and getting any of it to work locally is a project in…
Search
Top stories
I stopped opening Task Scheduler after discovering VS Code's hidden automation layer
4d agoI added this open-source tool to my local AI stack, and my local LLM finally has persistent memory
5d agoThe AI automation tool nobody talks about just replaced my entire workflow setup
6d ago6 projects I’ve already built with my $15 Raspberry Pi Zero 2W
6d agoClaude's scheduled tasks finally fixed what ChatGPT, Gemini, and every other AI tool got wrong
…The main difference is that Cowork gives you local file access while cloud tasks give you reliability! Here are some of the tasks I've automated with Claude so far Set it…
I used Obsidian wrong for a year, here's the folder setup that finally clicks
…Related How I set up a local wiki for my projects using Obsidian My secret weapon for project mastery My previous Obsidian setup A lot of manual work For the better part…
Ollama is still the easiest way to start local LLMs, but it's the worst way to keep running them
…I've been running local LLMs for a while now, and I switched away from Ollama a long time ago for most of my projects. And it hides the settings that would…
I used a local LLM to turn my NAS from a storage dump into something actually organized
…Having everything run locally on my NAS using Ollama and Llama 3 was non-negotiable. That became the start of a weekend project to turn my NAS from a storage dump into…
Home Assistant slows down over time, but there's a nuclear option most people don't realize
…Lastly, rebuild your automations using local scripts. Manually re-import your saved automation text files. Instead of building massive, messy, single-file automations, break recurring tasks into reusable, modular local scripts and…
I connected my local LLM to Home Assistant through MCP, and now my smart home manages itself
…Plus, I was able to trigger a bunch of HASS automations, modify scripts, and update integrations from LM Studio, so there’s a lot of utility in this project. Related I deployed…
I added one tool to my local LLM setup, and it finally stopped making things up
…My gpt-oss20B model ended up being more usable than expected, so I kept it in my rotations rather than discarding it as a side project. As powerful as local LLMs have…
Docker bypasses your firewall silently, and most home labs don't realize it
…You can write iptables rules there directly, or use ufw-docker , a community project that automates the integration between UFW and Docker's chains. It works well and is widely adopted, but…
I stopped trying to replace my cloud LLMs, and local models finally made sense
…Repetitive and automated tasks are another strong suit for local models. I wrote a Python script that feeds saved articles to a local Llama model overnight for summarization . This is a good…