Software Architecture, Homelabs and Videogames
A chatbot once sold a Chevy Tahoe for a dollar because someone asked nicely. Here is what it took to put a real prompt-injection filter in front of my local LLM stack — and the one-character trick that walks straight past it.
My ISP blocks inbound ports, so I stopped trying to let the internet in and had my server reach out instead. A full Cloudflare Tunnel setup with nginx and Docker — no port forwarding, no public IP, nothing to scan.
Open a browser on any machine and have your full VS Code IDE waiting — extensions, git, SSH keys, and full host Docker access via DooD, with a local Qwen3.6 for AI coding.
“One more model won't hurt.” That sentence has probably caused more home-lab incidents than expired certificates, broken Docker volumes, and mysterious YAML indentation combined. It certainly caused mine. My NVIDIA DGX Spark was already busy. It was running chat models, an embedding service, speech-to-text, text-to-speech, vector search, automation
James Bond was my first memory. Watching my son play IO Interactive's 007 First Light handed the franchise down a generation — THEIA, the bluff, a reimagined Moneypenny, and why the games remember the fun the movies forgot.
Ever since I watched Iron Man, I have wanted my own JARVIS. Not necessarily the flying armor. Although, to be clear, I would accept the flying armor. What I really wanted was the assistant: something that could listen to me, understand my occasionally drunk-sounding broken English, remember previous conversations, search
OData has been around since before microservices were cool — quietly powering enterprise APIs while everyone else was busy reinventing REST for the 47th time (and still adding /v2 to the URL like it's 2009). Then GraphQL shows up — all swagger and curly braces — promising perfect queries, no overfetching,
Do you still open Google Analytics every Monday to see if your website’s traffic went up? Yeah… same here — until I automated the whole thing. In this post, we’ll create a workflow that fetches your site metrics from Google Analytics, formats them into clean HTML, and emails them
If you’ve ever watched your GPU twiddle its thumbs between prompts, this one’s for you. In this post we’ll cover what vLLM is, why it’s fast, how to run it with Docker Compose, and how to test it with real calls. I’ll also show concrete
Why Quantization Matters Large Language Models (LLMs) are huge. Even a “small” 7B parameter model can chew up 14+ GB in FP16 (16-bit floating point). If you’ve tried running one locally without a beefy GPU, you’ve probably noticed your machine crying in pain—or worse, swapping memory like
