Atomic agents, o9 aligns AI for complex cross-functional planning
Curiously named o9 has pushed forward the software application development components and tools used in its Digital Brain platform by incorporating generative AI-powered Large Language Model (LLM) composite agents.
Hardly a newsworthy development in an age when every platform vendor (or indeed smaller toolset specialist) is grafting a degree of generative intelligence into its product offerings, the company points out that its work here sees it develop next-generation software agents to help planners execute complex tasks for technologies designed to serve integrated business planning.
Atomic agents
What’s happening here relates to o9’s composite agents, which are built on a foundation of atomic agents – these are AI-driven systems that are capable of performing tasks, retrieving information and generating responses to queries based upon specific inputs, coupled with LLM systems that are trained in information that is specific to a business.
Within the o9 Digital Brain platform’s core Enterprise Knowledge Graph (EKG), composite agents are able to integrate a sequence of atomic agents to perform more complicated exercises that can be used in cross-functional planning processes, such as building a forecast or creating a post-event analysis.
This is done through a sequence of retrieving and synthesising data to arrive at an outcome, such as analysing month-to-month forecast changes or providing a post-event analysis (of, for example, a previous quarter) to compare a prior forecast to actual results and perform root-cause analysis on forecast inaccuracies.
Agent recipes
Additionally, composite agents are trained on the ‘recipes’ followed by an organisation’s planning experts and can continuously learn from feedback.
The company insists that this results in agents getting better at arriving at the desired outcome.
“Our Digital Brain’s EKG offers cross-functionality as it touches supply chain, finance, procurement, commercial, customers and suppliers. It connects the knowledge of all these entities – and is constantly accumulating new knowledge as well,” said Anand Srinivasan, o9’s chief strategy officer. “Now we are deploying composite agents that can do a lot more cross-functional analysis.”
Srinivasan says that typically, an employee within a particular function only has the purview of knowledge, people, and systems within their department; they don’t look across and consider, ‘How does my decision impact someone else in a different department or organization?’
Hence, a lot of things get missed and there is a lot of value leakage. This is another thing these composite agents can help solve by becoming more cross-functional in their analysis.
“In the years to come, we believe that the quality of a business’s digital knowledge and how it informs the company’s business planning and decision-making will be a competitive advantage that propels it forward,” said Chakri Gottemukkala, Co-Founder and CEO of o9.
Gottemukkala notes that part of what makes today’s sophisticated AI systems so powerful and versatile is the focused expertise and teamwork between atomic agents and the holistic oversight of composite agents that can manage a series of tasks and integrate their outputs to achieve overarching business initiatives.
