Three weeks of building instead of writing. Here is everything that piled up in the meantime, including a few days with Anthropic’s new top model.

Three AI graphic novels, one research question: where does the line between slop and quality actually live? What changed across 255 agents, 16M tokens, and three complete pipelines.

A month ago I resubscribed to Codex because Anthropic's limits kept stopping me. Then came three 5.5 models in fifteen days, a Sonnet cache cut, and $200 of API credit tucked inside my Max plan.

Why homemade beats off-the-shelf, what actually makes an agent learn, and a job finder built for friends.

The quiet ways the agent you already run falls apart in use, and how to make fewer of them.

Ten months of running my agent from a black window, eight interfaces that almost worked, and the open source agentic IDE that ended the hunt this week.

It reads their CV, searches every morning across many sources, explains why each role fits, and gets sharper when they reply to the email. Here is the whole thing, written down.

The data was public all along. AI made the cleaning and joining cheap, so the fun question became what to cross with what.

Why the fallback layer is the cheapest resilience you can ship, what the $20 actually buys you, and the capabilities I deliberately keep off the primary stack.