DMSpace

A culturally informed, judgment-free space for reflection, emotional regulation, and connection.

Comment

Inspiration

This project was inspired by a gap I kept noticing between clinical mental health tools and the everyday emotional experiences people actually have. Many existing platforms either feel too clinical, too generic, or culturally disconnected from how users understand and express mental wellness. As a graduate student working at the intersection of technology, data, and health, I wanted to explore how an AI-assisted system could provide gentle, reflective support without positioning itself as therapy or crisis care. I was especially motivated by the idea that mental wellness is not one-size-fits-all. Cultural background, values, and lived experiences shape how people process stress, emotions, and growth. This project became an experiment in designing a space that feels supportive, adaptive, and human-centered, rather than diagnostic or prescriptive.

What it does

DMSpace is a culturally-informed mental wellness application that provides a safe space for users to reflect, journal, and connect through gentle AI-supported interactions. Users can chat freely, receive empathetic reflections, generate personalized journaling prompts, track moods, engage in wellness activities like breathing exercises and gratitude practices, and optionally explore peer-to-peer connections based on shared emotional themes. The app is designed to support self-awareness and emotional clarity without positioning itself as therapy or crisis care.

How I built it

DMSpace was built using Python and Streamlit for the frontend and application logic, with AI-powered reflections and journaling prompts generated via the OpenAI API. Session state is used to manage chat history, mood tracking, journaling entries, and peer-matching logic locally and privately. The system uses structured prompts to guide the AI toward empathetic, non-judgmental responses, while allowing users to select cultural perspectives that shape reflection style. Lightweight keyword-based theme extraction and scoring are used for optional peer matching, and wellness games are implemented as interactive, stateful components within the app.

Challenges I ran into

One of the biggest challenges was defining clear boundaries for the AI; ensuring it remained supportive and reflective without acting as a therapist or crisis service. Managing Streamlit’s rerun behavior, session state, and deployment differences between local and hosted environments also required careful debugging. Additionally, designing an interface that felt calm, readable, and accessible across devices while handling secure API key management presented ongoing design and engineering challenges.

Accomplishments that I'm proud of

I'm especially proud of creating a fully functional mental wellness app that integrates AI reflection, journaling, mood tracking, peer connection logic, and wellness activities in a single cohesive experience. The culturally adaptive reflection system, privacy-conscious design, and clean user interface demonstrate how thoughtful technical decisions can support emotionally safe interactions.

What I learned

This project deepened my understanding of human-centered AI design, particularly how tone, structure, and constraints shape user trust. I learned how to manage stateful interactive apps in Streamlit, securely integrate external APIs, and design features that balance technical capability with emotional sensitivity. Most importantly, I learned that small design choices can significantly impact how supported users feel.

What's next for DMSpace

Next steps include improving mood trend visualization over time, expanding culturally adaptive coping strategies, refining peer-matching logic, and exploring persistent but privacy-respecting data storage options. Longer-term, DMSpace could evolve into a research-backed platform for studying reflective AI interactions and culturally responsive digital mental wellness tools.

Built With

Share this project:

Updates