Cognizant Enhances Neuro AI Platform with Multi-Agent Capabilities
Cognizant is amplifying its Neuro AI platform, initially unveiled last year, by integrating advanced multi-agent capabilities. This innovative enhancement is designed to enable organizations to ideate, prototype, and test generative AI applications without the need for extensive coding.
Babak Hodjat, Cognizant’s Chief Technology Officer for AI, shared insights with VentureBeat about the transformative potential of this platform. Traditionally, creating generative AI applications involved direct intervention from Cognizant’s experts. However, with the new updates, enterprises can now utilize the Neuro AI platform independently, empowering them to host the service in-house and develop innovative applications tailored to their specific needs.
“Our clients expressed a keen interest in utilizing the platform themselves after seeing its potential,” Hodjat remarked. Many organizations began perceiving Neuro AI as a ‘factory’ for generating ideas to apply generative AI within their operations.
The multi-agent feature distinguishes Neuro AI from competing platforms. As enterprises increasingly turn to AI agents for solutions, Cognizant has strategically reconfigured its service to meet growing demand. The Neuro AI platform operates through four essential steps, each supported by pre-configured agents—these include the Opportunity Finder, Scoping Agent, Data Generator, and Model Orchestrator.
Functionally, Neuro AI acts as a virtual consultant, guiding clients through the application development process and offering a structured framework to follow. Upon initiating their use of the platform, users outline the challenges they want to address, prompting the Opportunity Finder to deploy agents that explore industry-specific use cases. Following identification of a viable use case, the Scoping Agent evaluates its impact across various performance indicators. Then, the Data Generator produces synthetic data to facilitate testing, and finally, the Model Orchestrator completes the application setup.
The Model Orchestrator employs a variety of agents to collaboratively construct the system. Hodjat explains, “Agents communicate with one another to assess what capabilities are required.” Each agent specializes in different areas, enabling a seamless dialogue to discuss project needs and solutions.
To develop the multi-agent orchestration system, Cognizant utilized the LangChain framework, which allows for flexibility in model integration. This setup accommodates both open and proprietary models, aligning with client preferences. “Our aim was to create an agnostic framework that could adapt to various client requirements,” Hodjat noted, underscoring the platform’s versatility.
Navigating a Competitive Landscape in AI Consulting
Cognizant’s initiatives in generative AI are part of a broader trend in the consulting industry, with the company recently launching an AI lab in San Francisco to bolster enterprise AI adoption. The competitive landscape is heating up, with firms like Accenture and AWS introducing platforms that assess AI readiness and promote responsible AI practices. Similarly, McKinsey & Company has developed a chatbot named Lilli to enhance its consulting capabilities.
As consulting firms increasingly carve out their niches in the rapidly evolving AI platform market, organizations like Cognizant are crafting solutions that cater to businesses seeking guidance on effectively leveraging generative AI. With established enterprise software providers such as Salesforce, SAP, and Oracle already offering tools for building AI applications, the demand for tailored consulting services is set to grow, marking an exciting time in the world of AI innovation.
