For decades, enterprise software has been designed primarily to record what businesses already know: customer information, financial transactions, employee activity, inventory, sales pipelines, and operational decisions. These platforms became the systems of record that organizations relied on to maintain a consistent version of the truth.
Artificial intelligence is beginning to challenge that model. The emergence of multi-agent AI systems could push enterprise software toward something more dynamic: systems of action that do not simply store information but interpret it, coordinate decisions, and initiate tasks.
The shift is significant because enterprise workflows rarely consist of a single question with a single answer. A customer-service issue, for example, may require reviewing account history, checking contractual terms, identifying a potential solution, updating a ticket, and communicating with another department. A single AI assistant can help with individual steps, but coordinating an entire workflow requires something closer to a team of specialized digital workers.
That is where multi-agent orchestration enters the picture. For Swaroop Borukar, a seasoned Product Manager building in the AI Infrastructure domain at Workday, this represents a fundamental change in how enterprise applications can be conceived.
