The successful adoption of Agentic AI requires real-time and dynamic access to business data. Here, Edward Funnekotter, Chief AI Officer, Solace, discusses how approaching data movement from an event-focused perspective is crucial for intelligent and adaptive Agentic AI frameworks to solve business issues that require intricate problem-solving.
The evolution of artificial intelligence continues at pace and at scale. In a flash, we’ve moved from early use cases such as Retrieval-Augmented Generation (RAG) and Large Language Models (LLM) to meet the master of multitasking, Agentic AI, where AI agents can act autonomously, make decisions, and take actions with minimal human intervention to achieve specific goals.
Agentic AI brings autonomous business value with contextual knowledge
Agentic AI goes far beyond simple question-answering on a subject for which an LLM has been trained. It is a software pattern that uses multiple LLMs and services, aka “agents,” to perform more complex tasks and reasoning autonomously. The evolution of agentic frameworks has seen a remarkable transformation. Initially, these systems were limited to rule-based tasks. They have now steadily advanced into sophisticated, multimodal agents.
These agents possess the ability to process and integrate information from diverse sources, including text, images, and audio. This multimodality empowers AI agents with reasoning capabilities that can interact in ways that can almost simulate human understanding. It is like an employee that is flexible and adaptable with a specific area of expertise.
For example, it could act as a customer support agent with the task – “for each new ticket, find similar tickets and answer product usage questions that are in the ticket. Then add your findings as a comment to the ticket.” This opens up the ability to tackle a huge spectrum of business challenges, from intuitive customer communications to real-time inventory decisions, fleet management, and more.
Unlocking Agentic AI success is not the flip of a switch
That said, getting out of the pilot phase and into everyday business applications is proving to be the biggest hurdle for any AI project. HBR has estimated that AI projects have a failure rate as high as 80%. In fact, according to an IBM study of over 8,500 IT professionals worldwide, limited AI skills and expertise, data complexity, and ethical concerns were cited as top barriers to AI deployments.
In addition, other studies show many projects fail to scale due to legacy architecture dependencies but also due to costs and performance in scaling something so complex and unstructured. Even when projects do get up and running, data quality, governance, security, and tech workflow integration hurdles remain.
Welcoming the multitasking AI agent – with data powered by an Event Mesh and Event-Driven Architecture
At the heart of these challenges lies a critical deficiency – the absence of real-time, contextual information flow. Traditional batch processing and static data models still in use by many organizations fall short of providing dynamic business environments where decisions, often that have to be made in split-seconds if you consider financial trading, are the make or break of trading opportunities.
An event mesh, underpinned by event-driven architecture (EDA), is the missing ingredient that promises to transform enterprise AI into a real-time, context-aware powerhouse. An event mesh is an interconnected network of event brokers that dynamically routes event-driven information between all kinds of applications and devices across environments and around the world.
Here’s where the event mesh shines for AI deployment. It provides the decoupling needed for rapid development and change, and it delivers on the event-driven architecture that allows for managing rate mismatch, supporting different applications with messaging patterns, and delivering the efficiency needed to scale horizontally and vertically.
When you apply the architectural pattern enabled by the event mesh across Agentic AI use cases, you essentially create a flexible, real-time data distribution network that enables various AI models to access and react to relevant data streams instantly.
See also: A Perfect Pairing: EDA and ChatGPT
Agent Mesh reduces information complexity
While an event mesh enables real-time data flow and dynamic routing across the enterprise, an agent mesh takes this further by introducing intelligent agents that can autonomously reason about and act on this information flow.
An agent mesh is a framework that allows you to build a network of AI agents overseen and controlled by a dynamic orchestration layer, allowing complex tasks to use multiple agents and bring their results together in a data management system. Agent mesh gateways allow access to this system for many different use cases, each with its own type of input interface and authorizations.
Essentially, organizations can enable truly autonomous Agentic AI systems that can manage requests to deliver the best results based on unstructured inputs, such as chats.
A strong AI foundation allows for a phased introduction
Best of all, an agent mesh is not intrusive to an organization’s existing application stack and Agentic AI framework. With its ‘plug-and-play’ style approach, organizations can start small with one or two use cases and then, over time, evolve the agent mesh by adding agents to increase its capabilities, as well as new agent mesh gateways to add further use cases and interfaces to the system.
There is always room for improvement… let Agentic AI pave a clearer path for future AI deployments
Then, with orchestration and built-in access control of all agents and actions in the system, one framework can be used and re-used for many use cases – be it a new order, a new support ticket, or even a question from a chatbot – each providing different interfaces and access control that is governed by enterprise-grade security.
In a landscape where AI technologies are rapidly evolving, the decoupled nature of an event-driven framework underpinning agent mesh allows organizations to easily update, replace, or add new AI models and data sources without disrupting existing systems. This is especially crucial for staying current with AI advancements.
No business blind spots – agent mesh shapes the future of autonomous systems
Timely access to up-to-the-minute data is a fundamental prerequisite for Agentic AI to deliver on its transformative potential. The agent mesh is vital to optimize the effectiveness of AI agents to meet the demands of dynamic business requirements.
Without that real-time, contextual information flow that the mesh provides, any Agentic AI system is essentially operating with blind spots. It can’t fully react effectively to changing business conditions or make truly insightful decisions.
