aiFlows
aiFlows is developed in EPFL’s Data Science Lab (dlab) under the supervision of Prof. Robert West. Originally a code-to-code translation tool leveraging feedback loops between LLMs, it evolved organically into a broader framework for defining interactions between AI agents and other agents.
Such collaborations between AI agents, non-AI agents and humans will become increasingly common in our daily lives to tackle complex tasks. For instance, an AI agent could assist a doctor in monitoring their patients by communicating with AI agents running on medical devices worn by their patients.
Structured interactions between autonomous agents are key for such collaborations. The aiFlows framework and its associated FlowVerse provide just that. Complex tasks are broken down into self-contained computational units – so-called Flows – that interact through a standardized message-based interface. This allows each Flow to maintain an internal state and thus autonomy while interacting with other Flows in a well-defined manner. Crucially, this means that agents can be distributed across different devices and locations.