Sim

Sim is an open-source AI workspace where companies build, deploy, and manage agents across IT, HR, marketing, and go-to-market functions. Today's launch, Sim Search, is an agentic search module that lets a team securely connect their credentials and integrations so that agents can operate on scoped projects with the same context a human employee would have. The pitch in the launch video is straightforward. Instead of every new agent starting from zero, Sim Search assembles an agentic knowledge graph over your connected data, then uses it to answer questions, pick up unfinished work, and execute tasks end to end (the demo walks through an agent finding a leftover ticket to issue a Brex card for a new hire, then applying internal spend policy to complete the setup). The launch matters because credential sprawl and missing context are two of the practical bottlenecks slowing agent adoption inside larger organizations. Sim ships with over a thousand integrations built in-house for this workflow, and the broader platform is an open-source AI workspace where teams build, deploy, and manage AI agents for their organization under shared identity, policy, and spend controls. That governance layer, combined with self-hosting and per-agent permissions, is what pushes Sim toward buyers who need auditability alongside autonomy. Sim is built by co-founders Emir Karabeg and Waleed Latif , and went through YC's X25 batch earlier this year . The company says its platform is already in use at customers including Rivian, Volkswagen, and EXP Realty, which gives Sim Search an immediate deployment surface among teams already running agents in production.