From 479 preemptions to serving 128k contexts out of a 64 GB disk cache — including the upstream bug I had to patch to get there.
The model felt slow, so I stopped guessing. Prometheus on my Unraid box scrapes vLLM, LiteLLM, and the GPU — and a 31-panel Grafana dashboard separates 'slow prefill' from 'KV cache exhaustion' from 'the engine is dead'.
Why I left Cursor for Paseo — a remote daemon that gives me cloud-class fire-and-forget agents while the model stays on my own box. How the setup works, how the native Android app beats Cursor's iOS-only mobile story, and why the local/cloud split is the whole point.
From 26 t/s to 33 t/s with a 2x context window — what actually moved the needle, and the tradeoffs that didn't show up in the benchmark.
What the 15 digits in an IMEI actually mean, how the Luhn check digit works, and a client-side tool on spanners.dev for generating and validating IMEIs and IMEISVs for testing.
Eight brand-new developer tools just landed on spanners.dev — cron parser, regex tester, JWT debugger, diff viewer, and more. Plus a bunch of shared improvements across the whole site.
A practical walkthrough of setting up Hermes Agent as a lightweight orchestrator that drives GitHub Copilot through a fully automated issue → PR → merge pipeline. Not a polished CI/CD system — a crude but effective dev loop that actually works (mostly).
How I went from zero analytics to full Tinylytics tracking across 9 repos — almost entirely hands-off — using Hermes Agent as the orchestrator and GitHub Copilot to write the PRs.
Not 100% sure when this started, but I eventually got annoyed by Visual Studio and ReSharper having a different keyboard shortcuts for potential code fixes. ReSharper is `Alt + Enter Visual Studio...
You can't use a 102 status code because Azure will always wait for a follow up code. In my code I changed it to a 202 Accepted instead. If you try and use a 102 in Azure, you will eventually get a [5...