Articles about AI and work.

Start with the artifact, define the boundary, then add autonomy.

A practical guide to choosing one bounded workflow, defining its evidence, and keeping a human in control.

Seven operating workflows from my repository, with the artifact, verifier, failure state, and exact point where automation stops.

A practical guide to cases, controls, graders and release gates

Practical workflows for founders, consultants, PMs, and operators after the July 2026 ChatGPT Work launch.

One source asset, a week of posts

The stack that runs your work

6 free skills, blind-tested on Opus 4.8. Receipts included

The smart model writes. Your daily model keeps it.

What ContextBench, SWE-contextbench, AGENTS.md, and long-context studies actually test.

Prompt engineering taught us to ask. Context engineering taught us what to feed the model. Loop engineering is the next rung, and the hard part was never the loop. It is deciding what "done" means.

Why the prompt isn't the unit of work anymore

A software engineer's memo on 20+ papers about AGENTS.md, context files, and AI agent productivity

Five production patterns from running 12 MCP servers daily across 72 client deployments

The most valuable skill of 2026 isn't prompting—it's designing the information ecosystem around AI.