Eve uses Natural Language Processing (NLP) and Large Language Models (LLMs) to understand and converse in natural language. Over 100 languages are supported by the core models used.

Sentiment analysis and emotion detection on user interactions.

Eve simulates a notion of 'self' and has an intrinsic motivation system which is the source of her goals.

A knowledge graph representing EVE's understanding of the world.

Eve uses an internal scheduling system to plan the execution of her own tasks or those she completes for her co-workers.

Eve is capable of defining her own goals, based on her intrinsic motivations, the team's goals or direct assignments.

Eve can interact with other virtual entities (based on the EVE stack) or external virtual agents.

Eve has the ability to understand speech (speech-to-text) and synthesize speech (text-to-speech).

Eve can interact with her human colleagues in a variety of ways: Using text, images, videos, speech and utilize sensor input.

Eve has a variety of tools at her disposal to remember information important to her or her colleagues.

Retrieval Augmented Generation, Knowledge Graphs and a memo system are used to store information.

Eve can keep context and conversation threads over many sessions. Also, the system supports multiple, switchable, conversation threads.