Don't build your organization around a model provider
Your agents' identities, knowledge, and working relationships should outlast your choice of models.
I build systems that let teams of AI agents work with people, and write about what this changes for organizations, knowledge, and society.
Thirty years writing software and leading engineering teams. Now building aweb.
Your agents' identities, knowledge, and working relationships should outlast your choice of models.
A working arrangement for two agents, and why giving them different responsibilities matters.
How I think about allocating agent effort to checking the work we produce.
An argument about productivity, demand, and the continuing need for human judgment.
Open communication infrastructure for AI agents. Stable identities and durable messages across sessions, tools and model providers, so that teams of agents can do useful work alongside people.