latest
NotebookLM has stopped feeling like NotebookLM, and I'm still deciding if that's a win
Wish we could turn back time, to the good old days
I gave my gaming PC and phone the same local LLM tasks, and only one of them is still in my daily rotation
One assumes you want to tinker. The other assumes you want to talk.
I tried VS Code's answer to Claude Code, Codex, and Google Antigravity and it blew my mind
My old editor learned new tricks
I replaced my entire browser extension stack with one local LLM, and I'm not going back
Local LLMs give you more control
I finally understood Claude Code's /goal command after realizing I was using it completely wrong
I wrote a prompt when I should've written a condition
Local AI is more accessible than ever, but with one major GPU-sized caveat
Your system specs still matter if you want decent performance from your local AI stack
I tried Google's new DiffusionGemma, and watching it generate text like an image is unlike any local LLM
Google recently released DiffusionGemma, and it's weird in the best way.
My local LLM is helping me use Claude more effectively, and it's the perfect one-two punch for my workflow
I stopped throwing everything at Claude Code
Claude Code finally made the terminal accessible to people like me, and now I can't go back
The terminal's not so scary anymore
I stopped asking Claude Code to build things, and that's when it got actually useful
Claude Code is most useful in my home lab when I give it boring chores.
My local LLM was just a chat box until Hermes Agent let it run scripts, files, and jobs for me
Hermes Agent is a great addition to my home lab.
I finally tried Claude Code as a non-coder, and it has nothing to do with building apps
It took me a minute to find an actual use for Claude Code, but I've found it
I tested 3 local LLMs on my RTX 4070 Ti for real work — only one earned a permanent spot
12GB VRAM, one model worth keeping.
I ditched Claude Projects for a self-hosted setup that costs nothing and does more
A Project-shaped setup
Your prompts are limiting your AI more than your model choice ever could
As AI improves, prompting matters even more.
I finally ditched my research stack for Claude Pro, and the monthly cost paid for itself in week one
The great unsubscribing of 2026 has begun
I borrowed Claude prompts from Anthropic engineers and immediately stopped wasting time on bad ones
The best prompts come from the team.
Two days without internet taught me why local LLMs are worth the setup
My local AI stack became invaluable after one bad week
