PTSDfy

Post-Traumatic Stress Disorder affects millions of people worldwide, but finding immediate and accessible tools to manage symptoms can be difficult. I built PTSDfy to help fix this.

Comment

Inspiration

The inspiration for PTSDfy came from recognizing that in moments of acute distress—like panic attacks or dissociation—the digital world is often too loud, cluttered, and overwhelming. I wanted to create a "digital sanctuary" that doesn't just provide information, but actively participates in a user's grounding process. My goal was to bridge the gap between clinical tools and immediate, compassionate support, ensuring that no one has to navigate a flashback or a "dark moment" alone.

What it does

PTSDfy is a privacy-first mental health companion designed to lower cognitive load and promote emotional regulation through:

AI-Powered Journaling: I integrated Gemini 2.5 Flash to provide non-judgmental, immediate validation. To ensure total safety, I designed it so all data is stored locally on the user's device.

Interactive Grounding: I built a guided Box Breathing tool and a 5-4-3-2-1 Sensory Wizard to help users reconnect with the present moment during dissociation.

Emotional Mapping: A mood calendar I developed that allows users to track patterns and triggers over time visually.

Crisis Safety Net: One-tap access to emergency resources like the 988 hotline for immediate professional support.

How we built it

I built PTSDfy using a modern, high-performance stack to ensure the UI felt fluid and "calm":

Frontend: I used React 19 and TypeScript for a robust, type-safe user interface.

Styling: I utilized Tailwind CSS v4 to implement my "Safe Space" aesthetic, focusing on soft sage greens and organic shapes.

AI Integration: I integrated the Google Generative AI SDK to leverage Gemini 2.5 Flash for rapid, empathetic responses.

Build Tooling: I chose Vite to ensure a lightning-fast development and build process.

State & Storage: I leveraged browser Local Storage to ensure that sensitive journal entries never touch a cloud database, keeping user data under their own control.

Challenges I ran into

One of the primary challenges I faced was balancing the AI's personality. I had to carefully prompt the Gemini model to be supportive and grounding without attempting to provide medical diagnoses. Another technical hurdle was UI/UX optimization for high-stress moments; I had to strip away unnecessary clicks and ensure the "Box Breathing" and "Crisis" buttons were accessible within milliseconds of the app loading.

Accomplishments that I'm proud of

Privacy-First Architecture: I’m proud of successfully implementing a system where personal reflections stay entirely on the user's device.

The "Calm" UX: I spent a lot of time on the visual design to ensure it feels genuinely soothing rather than clinical or cold.

Seamless AI Integration: Getting Gemini 2.5 Flash to act as a reactive, gentle companion that feels "human" in its responses.

What I learned

Through this project, I learned a great deal about Trauma-Informed Design. This meant understanding that bright colors, complex menus, or sudden pop-ups can be triggering. I also deepened my knowledge of the Google AI SDK and how to use local storage effectively for complex, persistent data without needing a traditional backend.

What's next for PTSDfy

Voice-Activated Grounding: I plan to implement "Help me" voice triggers for hands-free assistance during severe episodes. Wearable Integration: I want to explore connecting with heart rate monitors to proactively suggest breathing exercises when a spike in stress is detected. Guided Meditations: Adding a library of audio-visual nature scapes to further the "sanctuary" feel.

Built With

  • browser
  • date-fns
  • google-generative-ai-sdk-(gemini-2.5-flash)
  • lucide-react
  • react-19
  • tailwind-css
  • tailwind-css-v4
  • typescript
  • vite
Share this project:

Updates