Alexa+ AI Assistant

Inspiration

We wanted to explore how a conversational AI assistant could help people manage everyday activities through a single, easy-to-use interface. Tasks, reminders, and smart-home interactions are often spread across different tools. Alexa+ explores how these workflows can be brought together in one assistant experience.

What it does

Alexa+ is a web-based AI assistant prototype built around conversational interaction and everyday productivity.

The project includes task management, reminder management, and a simulated smart-home interface. Its backend exposes APIs for application functionality and supports persistent storage through SQLite for local development and Supabase PostgreSQL for the hosted environment.

The architecture also includes a Model Context Protocol (MCP) endpoint intended to support structured tool interactions. The exact capabilities available depend on the configured tools and the deployed implementation.

How we built it

We built the frontend with React, TypeScript, and Vite, and the backend with Python and FastAPI. The backend handles application APIs and database access. SQLite supports local development, while Supabase PostgreSQL is configured for cloud persistence.

The frontend is deployed through Vercel, and the backend is hosted on Render. We added database connectivity checks, automated tests, environment-based configuration, and deployment documentation to support reproducibility.

Challenges we faced

One major challenge was adapting the database configuration for local and cloud environments without disrupting the existing desktop application. We also worked through deployment configuration, database connection strings, API connectivity, and the limitations of free hosting services.

These challenges reinforced the importance of separating development and production configuration, testing database connectivity, and verifying the complete application rather than relying on deployment status alone.

What we learned

We learned how to connect a React frontend to a FastAPI backend, configure PostgreSQL for a hosted application, maintain local development support, and prepare an application for cloud deployment. We also learned that reliable reminders and other background tasks require special consideration when a hosting service can sleep during periods of inactivity.

What's next

Our next goals are to complete end-to-end verification of the hosted frontend and backend, strengthen the MCP integration, improve reminder reliability, and expand the assistant's capabilities based on testing and user feedback.

Note: Alexa+ is an independent prototype. Do not interpret the name as a claim of official Amazon affiliation, and do not treat simulated smart-home interactions as physical device control.

Built With

• fastapi
• modelcontextprotocol
• postgresql
• python
• react
• render
• sqlite
• supabase
• typescript
• vercel
• vite