Field notes on applied AI
Notes on agentic workflows, product design when software can run the work, and what still requires human judgment. Subscribe via RSS.
- If AI saves time, why does the workday keep expanding?AI makes more goals feel reachable. Why ambition can turn saved time into a heavier workload—and why that may be America's lasting AI advantage.
- 13 illustration styles, revisited: GPT Image 2 vs. Nano Banana 226 images, 13 styles, one shared scene: a side-by-side comparison of GPT Image 2 and Nano Banana 2, with notes on style, detail, and composition.
- The first hour and the next few monthsBuilding with AI can move faster than your understanding of the app. The long work is learning what exists, how it fits together, and what to ask for next.
- Design AI workflows like access can breakAI access is now a business dependency. Dependencies fail. Here is how to build workflows that survive.
- The uneven days of building with agentsBuilding gets faster. Deciding what deserves the time, finding customers, and making the economics work still take judgment.
- What changes when an agent can do the boring part?A proposed loan onboarding workflow, and how I would judge whether faster fixes create a better product and a stronger business.
- 11 illustration styles that work well with AI image modelsThe same scene recreated in 11 styles across GPT Image and Gemini, with the exact prompt blocks I used and notes on what held up.
- How Git worktrees workGit worktrees give each Codex task its own folder and branch without duplicating the repository.