Conversational AI Orchestration for Seamless Virtual Agent Handoff
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing virtual agent systems face scalability and modularity issues due to monolithic architectures, leading to increased testing and debug time, synchronization of release schedules, platform limitations, and data privacy concerns, while separate instances fail to communicate seamlessly during user sessions.
Innovation Solution
Implement a Conversational Artificial Intelligence (CAI) orchestrator that manages communication between multiple virtual agents, determining user intent and routing requests to the appropriate agent, enabling seamless transitions and data transfer across agents.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a monolithic virtual agent system is used, then the system can handle complex requests, but the system becomes difficult to maintain and scale with increased testing and debug time
Solution Approach 1:
The patent divides the monolithic virtual agent system into multiple independent virtual agents, each responsible for specific skills or functions. This segmentation allows each agent to be developed, tested, and maintained independently, reducing overall system complexity while maintaining the ability to handle complex requests through agent collaboration.
Solution Approach 2:
The patent introduces an orchestrator as an intermediary component that manages communication and coordination between multiple virtual agents. The orchestrator receives user requests, determines which agents are needed, and facilitates seamless interaction between agents, thereby reducing the complexity burden on individual agents while maintaining system versatility.
2Device complexity
If multiple separate virtual agent instances are used, then development and maintenance become easier, but seamless communication and data transfer between agents deteriorates
Solution Approach 1:
The orchestrator serves as a mediator that establishes standardized communication protocols and data exchange formats between separate virtual agent instances. It manages state information transfer and coordination, ensuring reliable and seamless communication while preserving the independence and manageability of individual agent instances.
Solution Approach 2:
The patent implements a universal communication interface and standardized data formats that enable any virtual agent to interact with any other agent through the orchestrator. This universal protocol ensures consistent and reliable communication across diverse agent instances while maintaining their individual manageability.
3Productivity
If virtual agents are developed with independent skill sets, then scalability improves, but synchronization of release schedules and platform limitations worsen
Solution Approach 1:
By segmenting the system into independent virtual agents with specific skill sets, each agent can be developed, tested, and deployed independently at different rates. The orchestrator manages the integration of these independently developed agents, eliminating the need for synchronized release schedules while maintaining system coherence.
Solution Approach 2:
The system adopts a dynamic architecture where virtual agents can be added, removed, or updated independently based on development needs. The orchestrator dynamically discovers and integrates available agents, allowing flexible deployment schedules without requiring platform-wide synchronization, thus improving scalability while managing complexity through adaptive orchestration.
Data Source
AI summary
Techniques are disclosed that relate to a computer system implementing a conversational artificial intelligence platform. The computer system may receive a request that includes text data specifying an utterance of a user. The computer system determines, form the utterance, an intent of the user. The computer system selects a first virtual agent to handle the request based on the determined intent. The computer system routes the text data and subsequent text data of the request to the first virtual agent for handling. The computer system receives, from the first virtual agent, a message that the determined intent is incorrect, the message including state information indicative of actions taken by the first virtual agent relative to the request. The computer system selects a second virtual agent to handle the request based on the state information. The computer system may receive input from an interactive voice response system or another channel.