Projects

Personal AI & AI Twin Systems

Designing systems where a person can have an AI-enabled layer for knowledge, communication, memory, and interaction.

This work explores how a person can have an AI-enabled layer that represents knowledge, communication, memory, and interaction over time. The core question is not just how to build a chatbot, but how to design a system that feels meaningfully connected to a real person and useful across different contexts.

Problem / opportunity

Most AI systems are generic. They respond, but they do not carry real continuity, personal structure, or a deeper relationship to identity and long-term knowledge. If personal AI is going to matter, it needs more than a prompt box. It needs a stronger foundation for ownership, context, memory, and interaction.

What the system does

These systems explore how individuals can shape their own AI layer for public interaction, private assistance, knowledge organization, and longer-term representation. The work includes both conceptual architecture and practical product thinking around how such systems should behave, evolve, and remain useful over time.

Examples of the kinds of problems this work addresses:

My role

My contribution has included product architecture, UX direction, concept design, system logic, documentation, feature definition, and iterative prototyping. I have worked especially on connecting abstract ideas around personal AI into clearer product structures and user journeys.

Product / UX perspective

The key product challenge is making personal AI feel trustworthy, coherent, and personally meaningful. A personal AI should not feel like a generic assistant with a name attached to it. It should reflect a deeper logic around what the person wants it to represent, what it should help with, and how it should evolve.

This also means thinking carefully about:

  • Onboarding and first-use experience
  • Role clarity for the AI
  • How a person understands and controls the system
  • What should feel conversational versus configurable
  • How the experience stays useful over time

AI / architecture perspective

This work requires careful thinking around:

  • Knowledge structures
  • Memory layers
  • Identity and representation
  • Conversation flows
  • The balance between structured and flexible information
  • Public versus private contexts
  • Long-term continuity

A core principle is that personal AI is not only a model problem. It is also an identity problem, a memory problem, a UX problem, and an architecture problem.

Key design lessons

This area continues to interest me because it combines human-centered design, system architecture, AI interaction, and long-term product thinking in a way that few product categories currently do.

← All projects Contact