Em Shreve describes two-week, in-person learning cohorts for a fully distributed company. Half the time is facilitated learning and half is spent on participants' real projects, with peer workflow sharing and daily demonstrations. Separate programs address engineers, cross-functional groups and nontechnical teams.
The engineering curriculum progresses from skills, MCP useModel Context Protocol is a standard way for AI applications to connect with external tools and data sources through consistent interfaces. and responsible AI to harness extensionsAn AI agent harness is the software framework that packages a model with tools, instructions, context management, execution controls, and user interaction. and autonomous routines. Em Shreve emphasizes updating the curriculum as tools change. Shared MCP infrastructure and a portable skills directoryAn AI agent skill is a reusable package of instructions, resources, and tool guidance for performing a bounded kind of work. let cohort projects become reusable company capabilities.
Participants are encouraged to become teachers through a guides program, regional gatherings, workflow ride-alongs and an ongoing help channel. Follow-up surveys at 72 hours, 30 days and 90 days track intentional use, integration and confidence. Examples include store analytics, support connectorsApplication integration connects software systems so data and actions can move through one useful workflow. and guided admin assistance, illustrating why meaningful learning time matters.
Watch on YouTube
