Abstract
Effective Artificial Intelligence (AI) Agent design is challenging, usually done in isolation by a technical resource and requires a lot of overhead configurations, in addition to dealing with the complexity of an AI Agent designer that requires programming along with slot filling (collecting data for backend external system integrations, called fulfillments). When used in a dynamic contact center environment, an increasing number of AI Agents are required for varying patterns including load, agent skillset, high traffic, and newer contact topics. Conventional design of AI Agents uses analysts to understand bot efficacy to tailor the design for improved customer experience. To overcome these issues, proposed herein are techniques leveraging the advances in artificial intelligence, specifically large language models (LLMs) that can understand conversational transcripts and existing Application Programming Interface (API) documentation using prompt engineering, to generate AI Agents (bots) on demand, in a self-created manner.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
R Damodaran, Vinay; Asthana, Aseem Banshidhar; P Chotai, Ashish; and Bhattacharjee, Arunabh, "ADAPTIVE SELF-GENERATING AI AGENTS WITH AUTOMATIC EXTERNAL SYSTEM FULFILLMENTS", Technical Disclosure Commons, (September 30, 2024)
https://www.tdcommons.org/dpubs_series/7386