AI development, engineering management, communities.
Spec-driven development without spec sprawl: how to keep specs in one aggregate dictionary so they do not multiply with every feature: requirement levels, the aggregate card and context boundaries.
Four kinds of persistent context for AI (AGENTS.md, CLAUDE.md, skills): the task spec, the project bible, rules and skills. What goes into each, when it enters the context, and what ChatGPT and Claude offer. Based on Anthropic, OpenAI, GitHub, Cursor and Cline docs.
How to pick 16 people from the report, a questionnaire and license lists, check team coverage and replacements, and decide what to ask them.
Prompts for work with an AI chat: the parts of a good work prompt, what a weak and a strong ask look like, six ready templates for common tasks and how to keep them in a project, based on the Anthropic, OpenAI and Google guides.