A new open-source project called OpenExecutive presents companies with a single AI executive persona backed by eight specialist agents, each built on Anthropic's Claude models, according to the project's GitHub repository.
The system, built by SenteLabsAI, keeps what it calls episodic memory of past decisions across sessions and combines built-in business knowledge with documents a company uploads, using ChromaDB and SQLite for storage. It covers strategy, finance, HR, legal, operations, marketing, product and board communications, and can be reached through a web interface, Slack, email, Discord or Telegram. The internal multi-agent structure is never exposed to the user, who interacts with what looks like one coherent executive. The repository is licensed under Apache 2.0.
The project climbed to the top of Hacker News this week under the title "CEO fired developers to make room for AI. Developers create open source AI CEO," drawing 787 points and 536 comments. The title's premise, that a specific chief executive fired developers to replace them with AI, comes from the post itself; Poliety found no independent account identifying which company or executive it refers to.
Whatever prompted it, the tool is real and usable: a free, self-hostable framework that gives a company a functioning executive layer without hiring for eight distinct roles, built on the same models many of those companies already pay for through Anthropic's API.