Senti App

Senti: Detect Early, Live Fully

Comment

Senti: Detect Early, Live Fully

Inspiration

The idea for Senti emerged from our shared passion for mental wellbeing and the desire to create a positive impact in our community. My sister and I, both deeply committed to mental health, recognized that early detection of emotional distress could prevent crises and empower people to live healthier lives. This personal drive, combined with our interest in innovative technology, inspired us to harness AI as a tool for compassionate, proactive support.

What It Does

Senti integrates AI to analyze user behavior—ranging from daily physical activities, sleep patterns, and digital interactions with other apps—to detect early signs of emotional distress. It provides custom affirmations, personalized activity recommendations, and a conversational interface that offers a safe space to express feelings. Additionally, Senti connects users with local community resources, helping them build a supportive network and engage in activities that enhance their wellbeing.

How We Built It

  • Technology Analysis & Architecture: Before development, we conducted an in-depth analysis of the most suitable technologies for Senti. We chose:

    • Frontend: JavaScript with React Native for a seamless, cross-platform mobile experience.
    • Backend: Python with FastAPI for handling AI-driven data processing.
    • Machine Learning: PyTorch and OpenAI models to power sentiment analysis and personalized recommendations.
    • NLP & Sentiment Detection: NLTK to enhance language understanding and emotional tracking.
    • Security & Privacy: OAuth 2.0 for secure user authentication.
    • Cloud Infrastructure: AWS to scale efficiently and handle AI model deployments.
    • External API Integrations: Google Health Kit for behavioral data tracking and Google Maps API for local activity recommendations.

  • Focusing on the Prototype: Given the time constraints of the hackathon, we prioritized designing an intuitive, user-centric experience in Figma. Our prototype illustrates:

    • AI-driven insights and real-time mental wellness tracking.
    • An interactive conversational interface for self-reflection and motivation.
    • Localized recommendations for activities and support networks.

Challenges We Ran Into

  • Balancing Technology vs. Design: While we defined a robust technical architecture, we had to focus on designing the most impactful user experience rather than full implementation.
  • Ensuring AI Relevance: Fine-tuning AI models to detect emotional distress accurately while avoiding false positives or negatives remains an ongoing challenge.
  • Privacy Considerations: Given the sensitive nature of mental health data, ensuring security and transparency in user data handling was a primary focus.

Accomplishments That We're Proud Of

  • Strategic Tech Planning: Defining a scalable architecture that integrates AI, security, and behavioral analytics.
  • User-Centric Design: Creating an intuitive, aesthetically appealing prototype in Figma to visualize the user experience.
  • Innovative AI-Driven Approach: Implementing sentiment analysis and behavioral tracking to offer personalized, proactive mental health support.
  • Community-Centered Focus: Integrating Google Maps API for real-world activity recommendations and local mental health resources.

What We Learned

  • The Importance of Early Intervention: Behavioral changes can signal emotional distress, reinforcing the value of proactive detection.
  • Balancing AI & UX: A well-designed interface is just as critical as AI accuracy—users must feel comfortable and engaged
  • Technical Feasibility: AI-powered mental wellness apps require a delicate balance of accuracy, privacy, and user adoption.

What's Next for Senti App

  • Bringing the Prototype to Life: Developing the MVP based on the React Native + FastAPI + PyTorch tech stack.
  • Expanding AI Capabilities: Enhancing sentiment analysis and refining AI-driven recommendations.
  • User Testing & Feedback: Validating our prototype with real users to improve engagement and trust.
  • Integrating with Wearables: Connecting Apple Watch & Fitbit data for even richer mental health insights.
  • Building Partnerships: Collaborating with mental health organizations and local communities to expand support networks.

Senti is more than an app, it’s a commitment to proactive mental health support, combining AI-driven intelligence with human connection for a healthier, more mindful life.

Share this project:

Updates