Aria - Voice AI Agents
A multilingual, context-aware voice AI agent designed to have natural conversations, remember context, and take real-world actions.
- My Role
- Product Designer & Builder
- The Problem
- Most voice assistants are built around simple commands and one-off interactions. They can answer questions or perform basic tasks, but they lack deeper context, long-term memory, and meaningful personalization. Users still have to repeat information, switch between apps, and manually manage the next steps.
Overview.
I explored how voice AI could move beyond traditional assistants and become a persistent, multilingual agent that understands users over time. The system combines real-time voice interaction, AI reasoning, long-term memory, and external tools to help users complete tasks through natural conversation.
As the product designer and builder, I worked across the experience and technical architecture, exploring how voice, memory, multilingual conversations, and tool integrations could come together into a scalable AI agent. The goal is to create an assistant that feels less like a command-based interface and more like an intelligent layer that can understand context and act on the user’s behalf.