What we're writing about.
Notes from the studio — development, design, AI, industry shifts, career growth, and life around the team.
Browse the archive.
Kimi K3 Is Here — And It's Chasing the Frontier at a Third of the Price
Moonshot AI's Kimi K3 is the largest open-weight model ever shipped. We break down how it stacks up against the closed US flagships it's chasing — OpenAI's GPT-5.6 Sol and Anthropic's Claude Fable 5 — on benchmarks, price, and openness.
Understanding Nginx: Architecture, Internals, and Configs Beyond the Basics
What Nginx actually is, why its event-driven architecture makes it fast, and the advanced configuration patterns — reverse proxying, load balancing, caching, rate limiting, and API-gateway tricks — that separate a copy-pasted config from one you truly understand.
Tailscale: How I Reached My Mac mini Without Exposing a Single Port
Tailscale builds a private, encrypted mesh network between your devices — so you can reach a machine at home from anywhere without port forwarding, a static IP, or opening your firewall.
Ghostty and libghostty: The Terminal Core Quietly Reshaping the Ecosystem
I went from the stock macOS Terminal, to Ghostty this year, to Supacode this week — and only then realized the last two run on the same engine. Here's the story of Ghostty, libghostty, and why a reusable terminal core matters.
Pi: The Agent Harness That Defers, On Purpose
Earendil's Pi is a minimal coding agent that ships primitives instead of features — deferring MCP, sub-agents, and plan mode to you, your extensions, and the tools you already run.
GrapheneOS, Explained: How a Hardened Android Rethinks Mobile Security
A plain-language tour of GrapheneOS — the privacy- and security-focused Android variant for Pixel phones — and the design ideas that make it different.
Homebrew 6.0.0 Is Here: Tap Trust, Sandboxing, and a Faster Default
Homebrew 6.0.0 lands with a new tap-trust security model, Linux build sandboxing, the faster JSON API on by default, and a clear sunset timeline for Intel Macs.
The Agent That Grows Its Own Skills: Why I'd Pick Hermes Over OpenClaw
OpenClaw made it normal to drop a stranger's SKILL.md into your agent — and the security bill came due fast. Here's why I prefer Nous Research's Hermes Agent, which writes, refines, and prunes its own skills instead of trusting a marketplace.
Introducing ExpressionEngine 7.4
Significant updates have been made to ExpressionEngine 7.4, such as improved member management, the addition of new CLI commands, updates to channel forms, and the integration of Redactor X as a new Rich Text Editor (RTE) field type.
Have a project in mind?
The same team behind these articles builds production platforms every day. Tell us what you're working on.
Let's connect → [email protected]