I Used 50% of my Codex Sub in 36 Hours
I didn't get faster by chatting faster. I got faster because I stopped juggling every Codex session myself and let Claude dispatch a crew of them instead.
I didn't get faster by chatting faster. I got faster because I stopped juggling every Codex session myself and let Claude dispatch a crew of them instead.
I set out to test a harness I had built. What I found is that the system you wrap around an AI can matter more than the AI inside it. Short answer: The model gets the headlines, but the harness — the system you build around the model — quietly decides what
A plain-language explanation of AI agents as the relationship between a model's brain, a defined job, and the app surface where work gets done.
Storyweaver showed that AI writing tools are most useful when they preserve context and help the writer stay in flow.
GitHub stopped feeling like developer territory once AI made branches, commits, and diffs useful as a safety net for learning in public.
Working with AI is less about vibe coding and more about operating a capable system with context, constraints, judgment, and durable workflows.
The rough first version of AgentShelf mattered because it turned years of Salesforce automation, flow design, and LLM experimentation into a working product direction.
Plain Text begins with the path from English degree to Salesforce admin to AI engineer, and how words became the bridge into building software.