Inspiration
Mental health struggles are common among young people, yet many find it difficult to talk about their emotions due to stigma, fear of judgment, or lack of access to support. Across different cultures, expressing feelings like sadness, anger, or emotional exhaustion is often discouraged or misunderstood.
We were inspired to build TheraMood after recognizing how hard it can be to even start a mental health conversation. We wanted to create a tool that feels approachable, empathetic, and safe — something that meets users where they are emotionally and helps them reflect without pressure or labels.
What it does
TheraMood is a therapist-style, emotion-based AI chatbot designed to support youth mental well-being.
Users begin by selecting an emotion from visually distinct, colored cards such as Sad, Angry, Overwhelmed, Depressed, Happy, or Describe for custom emotions. Once an emotion is selected, the chatbot adapts its conversational style and engages the user in a guided, therapeutic dialogue.
The chatbot:
Uses empathy and emotional reflection
Asks gentle, open-ended questions
Encourages self-awareness and healthy coping
Avoids diagnosis or medical advice
Provides a safe, judgment-free space for expression
TheraMood does not replace professional therapy but helps users explore and understand their emotions in an accessible way.
How we built it
TheraMood was built as a web-based application with a focus on simplicity, accessibility, and ethical AI use.
Frontend: React / Next.js with Tailwind CSS for a clean, emotion-focused UI
Backend: API-based architecture to manage emotion state and chatbot logic
AI Layer: Large Language Model (LLM) used with carefully designed system prompts to simulate therapist-style conversations
Design Approach: Emotion-based entry point, minimal friction, privacy-first (no user accounts required)
Special attention was given to prompt design to ensure the chatbot follows therapeutic conversation patterns such as active listening and CBT-inspired questioning.
Challenges we ran into
One of the biggest challenges was balancing realism with responsibility. We wanted the chatbot to feel genuinely therapist-like while ensuring it did not diagnose, give harmful advice, or claim to replace professional care.
Other challenges included:
Designing prompts that remain empathetic over long conversations
Handling sensitive emotional input safely
Structuring conversations to feel natural, not scripted
Keeping the interface simple without oversimplifying mental health
Accomplishments that we’re proud of
Creating an emotion-based UI that lowers the barrier to emotional expression
Designing a chatbot that feels reflective and supportive rather than generic
Implementing ethical safeguards and crisis-aware responses
Building a culturally adaptable tool that avoids clinical labeling
Completing a meaningful mental health project within a hackathon timeframe
What we learned
Through this project, we learned:
How impactful empathetic design can be in mental health tools
The importance of ethical boundaries when using AI in sensitive domains
How careful prompt engineering can shape AI behavior significantly
That mental health solutions don’t have to be complex to be effective
We also gained deeper appreciation for the responsibility involved in building technology that interacts with people’s emotions.
What’s next for TheraMood
Future improvements for TheraMood include:
Multilingual support for broader cultural accessibility
Cultural tone customization for different regions
Optional mood tracking over time
Voice-based interaction for accessibility
Collaboration with mental health professionals for further refinement
TheraMood has the potential to grow into a powerful, responsible digital companion that helps normalize emotional conversations and reduce mental health stigma among youth worldwide.
Log in or sign up for Devpost to join the conversation.