Luna

🌙 Inspiration

The idea for Luna was born in the wake of the 2024 presidential elections in Senegal, during which over 2000 individuals were detained amidst intense political tensions and civil unrest. This situation exposed the hidden mental toll that political instability can take on both individuals directly affected and the broader population. Many were left feeling anxious, isolated, and unheard. Luna aims to provide a comforting, judgment-free space to support mental well-being in such moments of crisis — and beyond.

🤖 What it does

Luna is an AI-powered chatbot designed to provide emotional support and empathetic conversation 24/7. It helps users dealing with stress, anxiety, loneliness, or trauma by offering:

• A safe, anonymous space to talk

• Motivational and compassionate responses

• Mindfulness and breathing exercises

• Non-judgmental companionship in difficult times

🛠️ How we built it

• Frontend: A responsive web interface built with React.

• Backend: A secure API layer using Node.js and Express.

• AI Integration: Leveraged OpenAI's GPT-4 API for natural and empathetic conversation flow.

• Security & Privacy: No user accounts, no data tracking — full anonymity.

• Deployment: Hosted on a secure cloud platform (e.g., Vercel or Heroku) with end-to-end encryption.

⚠️ Challenges we ran into

Tone calibration: Ensuring that the chatbot remains consistently empathetic without sounding generic or automated.

• Crisis detection: Designing it to identify red-flag terms (like suicidal ideation) and provide appropriate suggestions without crossing ethical lines.

• Privacy concerns: Implementing true anonymity while still providing meaningful support.

• Cultural sensitivity: Tailoring the chatbot’s responses to be relevant and respectful across different sociopolitical contexts, especially post-crisis regions.

🏆 Accomplishments that we're proud of

• Creating a functional, anonymous mental health chatbot in under 48 hours.

• Designing an intuitive, calming interface that encourages open conversation.

• Managing to reflect real-life empathy and cultural relevance within AI responses.

• Turning a painful national event into a project of healing and support.

📚 What we learned

The importance of emotional intelligence in AI applications.

How to build tech that is useful not just globally, but locally and contextually.

That even simple tools — when designed with empathy — can have a deep impact.

How to prioritize mental health while respecting user privacy and safety.

🚀 What's next for Luna

Multilingual support (starting with Wolof and French) to reach more local users in Senegal and West Africa.

• Integration with real mental health professionals for critical escalation paths.

• Mobile app version for more accessible, on-the-go support.

• Community features: anonymous peer support and group conversations.

• Offline mode for regions with limited internet access.

Built With

• ai
• frontend:-react
• gemma
• llms
• local
• model:
• sqlite
• tailwind-css-backend:-next.js-(node.js)
• typescript
• via