Switchroom grew out of work we knew firsthand. At Sourcewise, that work was industrial procurement: understanding a requirement, finding relevant suppliers and deciding what deserved a closer look.
It gave us a practical starting point for building procurement and sourcing agents. We knew the questions behind a search, the context a useful answer needed, and where a person’s judgment mattered.
The knowledge came first.
A supplier search carries more than a product name. There are requirements to understand, details to compare and information that may be incomplete. Experience in the industry helps you recognize what is relevant and what still needs checking.
That understanding shaped the way we approached AI: begin with the work, connect the information it depends on, and make the result useful to the person making the decision.
A capability companies could have themselves.
The next step was to bring that approach into other businesses. A company has its own knowledge, systems and ways of working. Its teams should be able to use AI around that context.
That became Switchroom: private AI infrastructure and agents built around the company using them. Procurement was the starting point. Engineering, operations and the work between teams open up further possibilities.
Connected to the business as it works.
The aim is to work with the company’s own information, including live data from the systems it connects. An order’s current status, an approved document or an updated requirement can become part of the context for a task.
Each workflow needs the right sources, access and checks. The connection should make clear where information came from and how current it is, so the team can understand the result.
Sourcewise gave us the starting point. Switchroom takes that practical knowledge into a broader question: what could your people do with these capabilities inside their own business?