SnapLogic today added an ability to create artificial intelligence agents, assistants and applications using enterprise data that is aggregated using the company integrated platform-as-a-service (iPaaS) framework.
That capability makes it possible to use SnapLogic Agent Creator to create, for example, AI agents across a wide range of tasks to help manage workflows in real time, says SnapLogic CTO Jeremiah Stone.
The overall goal is to make it simpler for IT organizations to take advantage of agentic AI technologies to replace much less flexible approaches to automation based on robotic process automation (RPA) and optical character recognition (OCR) technologies, he adds. “Agentic AI makes those technologies obsolete,” says Stone.
The SnapLogic iPaaS is already widely used by many organizations to create workflows spanning multiple applications. The company is now extending that capability via a SnapLogic Agent Creator in a way that makes it possible to securely access enterprise data, notes Stone.
That approach makes it simpler for organizations to build AI agents that can be trained to manage tasks across a wide range of processes, all of which can be centrally governed and integrated into a larger workflow.
Those AI agents will also be able to continuously improve as they dynamically iterate through multistep tasks or are exposed to more advanced reasoning capabilities enabled by the latest update to a large language model (LLM), says Stone.
It’s not clear to what degree AI agents will transform workflows, but humans will still be needed to supervise processes. Much of the manual toil previously required to manage those processes, however, is going to be increasingly automated in a way that will make organizations much more efficient, adds Stone.
It’s not clear who within organizations will be leading those efforts, but ultimately it will require the expertise of IT professionals working closely with subject matter experts to identify tasks that might be better handled by AI agents rather than humans. Those teams will then have to re-engineer workflows in a way that makes it possible to incorporate those AI agents.
Additionally, AI agents should make it possible for organizations to create new processes that drive services at levels of scale that previously it would not have been economically feasible to provide, notes Stone.
One way or another, AI technologies are being quickly democratized in a way that should reduce the current dependency organizations have on data science experts. In fact, there will soon come a day when organizations will first assign tasks to an AI agent rather than an employee.
In the meantime, savvy organizations should be identifying manual tasks today that might be better handled by AI agents. In most instances, they will find that the humans performing those tasks don’t especially enjoy performing them. Ultimately, the goal should be to enable humans to spend more of their time adding value to the business by engaging customers, rather than focusing on routine back-office tasks that, in the final analysis, don’t provide much differentiated value to the organization.
