Dot

Unlike conventional AI assistants that reset with every conversation, Dot remembers everything users share across sessions, building an evolving, personalized model of the user's life, preferences, goals, and relationships over time.

The memory architecture at the core of Dot is what distinguishes it from chatbots and general-purpose AI assistants.

As users share information through natural conversation, Dot constructs what it calls a theory of mind a continuously updated portrait of the user's personality, interests, routines, and important life context.

This portrait informs every subsequent interaction, allowing Dot to ask more relevant follow-up questions, make more personalized recommendations, and notice patterns across the user's shared experiences.

Under the hood, Dot dynamically routes queries to the most appropriate AI model for each specific task, drawing on seven to ten different large language models from providers including OpenAI, Anthropic, and Google simultaneously.

This multi-model architecture means users benefit from the specialized strengths of multiple AI systems without needing to choose between them, while Dot's memory layer maintains continuity and personal context across all model interactions.

Practical use cases for Dot include remembering important dates and personal commitments, providing a non-judgmental space for daily reflection and emotional processing, offering personalized recommendations for recipes, travel, books, and entertainment based on known user preferences, and helping professionals and students organize complex tasks and long-term projects.