Abstract: The systems and methods disclosed herein relate to a persistent memory-based architecture for avatar-based applications that maintains user-specific data across multiple interaction sessions. A briefing data that includes interaction data between a user and an artificial intelligence agent and/or user time-series data is accessed from a persistent memory. In response to receiving additional interaction data, portions of the additional interaction data are linked to the briefing data structure. A user-specific input is constructed by combining the briefing data structure, the portions of the additional interaction data, a predicted state output and/or a dataset retrieved from one or more data repositories. The user-specific output is transmitted to input layers of the artificial intelligence agent to generate a user-specific output. The briefing data structure is updated to include a representation of the user-specific output.

Abstract: The systems and methods disclosed herein provide an architecture for deploying artificial intelligence (AI) avatars that generate audio and video responses to user interactions across various infrastructure environments. The systems and methods disclosed herein establish connections between client applications and a first infrastructure layer that invokes multiple agents trained on domain-specific knowledge bases to retrieve associated data records. The retrieved data and user input are routed to a second infrastructure layer hosting a generative AI model that identifies relevant data fields and generates avatar responses as audio or video streams. The systems and methods disclosed herein are enabled to evaluate the user inputs and/or avatar responses by applying one or more validation criteria. The systems and methods disclosed herein adjust the user inputs and/or avatar responses based on the evaluation.