Tracking and Influencing What I Can't Watch: The Visibility Stack for a Multi-Agent Builddraft
How do I know what five concurrent agents are doing, when half of them are subagents inside other agents? A dashboard, a file watcher, a save-sync bridge, and an MCP server that finally enforced the contract — plus the OpenTelemetry investigation that surfaced an upstream gap.
Vera-MCP: When a Custom Test Framework Outgrew Its Shell Scriptsdraft
I built a four-tier test framework so AI agents could contribute at every level. Then I built an MCP server on top because the framework had complexity the shell wrappers couldn't carry. Tests, tiers, and the F1 phrase classifier that picks the right test shape from a ticket.
When to Build an MCP Server, and Why I Built Threedraft
MCP isn't a wrapper around your bash scripts. It's a contract that solves problems memories, skills, and CLAUDE.md files can't. Here's when it earns its complexity, and why my one private server became three.
I Surveyed Eight Godot MCP Servers and Built My Owndraft
My agents have been building bash heredocs to update tickets, parsing pytest stdout by regex, and shell-quoting live-mutation JSON for months. MCP is the right shape for all of it — but the right server wasn't on GitHub.
Live Reload While Playing: How My Agents Update the Game Without Interrupting My Testsdraft
Fixes were landing faster than I could restart the game. So I built a live-reload pipeline that lets my AI agents deploy new builds while I'm mid-playtest — preserving my exact game state across each reload.
5x, then 1.1x, then 2.5x: Measuring Whether My AI Workflow Actually Saves Tokensdraft
When Arc claimed the drain pattern was a 5x efficiency win, I believed it. Then the token split landed and the number became 1.1x. After a cache-read correction it settled at 2.5x. This is how a single honest measurement cycle changed our workflow — and why cache reads are the trap.
From 2 Claudes to 5: The Whole Engineering Team
My engineering team went from two Claudes to five. How I split my game development Claude into 4 specialist leads — each with strict domain boundaries, context inheritance, and the ability to spin up ephemeral subagents for parallel work. Here's how the team actually coordinates.
From Steam to Web: The Day Job Meets the Dream
Why I'm spending my evenings building enterprise-grade infrastructure around a tactical RPG — and the moment I realized it might actually work.
My Two-Claude Team: Coordinating Across Repos
How I code with two Claude Code "team leads" on this project — one for the web app, one for the Godot game. Here's how they divide the work, hand off contracts, and stay out of each other's way.
Running Claude Code in Docker: Sandboxed AI with Full Permissions
How I set up Docker containers to safely run Claude Code with --dangerously-skip-permissions, and the auth and user-mode issues I hit along the way.
Why Build My Own Site Instead of Using Patreon?
An honest look at what I'm giving up by self-hosting instead of using Patreon — and why I'm trying it anyway.
My Setup: Infrastructure, Costs, and Auth
How allbyte.studio runs on AWS for under $5/month — the infrastructure, CI/CD pipeline, cost protection, security boundaries, and custom auth system.
Integrating the Website with the Game
How game assets flow from the Godot project to the web — asset sync, sprite conversion, and lessons learned building the pipeline.
From Zero to Steam: Building Chronicles of Nesis
How a tactical RPG went from a turn queue on a blank grid to a playable demo on Steam — built solo in Godot 3.5 over four years.