AI for Good hackathon project

v1
v1
v1
localhost v2
v2
localhost v2
DataAnalysisInV2
Hospital Location on v2
red icon is symbol of blood donation, click on right blood icon to open chatbot search interface
Prompts: FIND BLOOD BANK, blood bank, blood center, blood donation center, I need blood bank
Version 1 feature
api
v1
V2
v2
Update

Introduction

MindMend, also known as Lovable Cursor Whisper, Whisper, or WellnessAI Silk, is an empathetic AI companion designed to understand and respond to users' emotions. It gently guides individuals through thoughtful journaling and meaningful conversations, offering a supportive space for emotional well-being. When urgent help is needed, this versatile chatbot seamlessly connects users to nearby blood banks, ensuring timely access to critical resources. Together, these four identities represent a unified mission: to provide compassionate support and practical assistance whenever it matters most.

Inspiration

We envisioned a compassionate space where thalassemia patients can feel heard and supported without judgment—amplifying human empathy through responsible AI. Our vision began with the need for a safe, judgment-free space where anyone could openly talk, reflect, and truly feel heard. We recognized that AI has the power to scale this kind of empathetic support, but only if it prioritizes privacy, consent, and genuine care. Inspired by the challenges faced by individuals with conditions like thalassemia—who often require urgent access to blood support—we built this chatbot to not only provide emotional guidance but also to bridge the gap between users and nearby blood banks. By combining compassionate conversation with critical healthcare connections, we aim to empower users with both emotional and practical support, whenever they need it most.

What it does

MindMend (lovable-cursor-whisper) allows users to journal daily, analyzes their emotional tone using advanced NLP techniques, and chats with them like a supportive therapist—while also providing vital information about nearby blood banks.

Journaling & Chat: Daily prompts and free-form conversations designed to feel empathetic and supportive—not clinical.
Crisis Utility: Quickly identifies blood banks and blood centers near the user, sharing contact details for urgent assistance.
Gentle Guidance: Offers grounding exercises, cognitive reframing, and practical action plans—without making diagnostic claims.
Emotion Insight: Uses in-house NLP models and api's to analyze tone, summarize moods, and detect trends to build personalized emotional support logic.

How we built it

1-Frontend: React + Tailwind CSS

2-Backend: Node.js + Express (v2); Python + FastAPI 3-AI: GPT-4o for conversation & emotion analysis; JSON

Built with React and Tailwind on the frontend, Node.js and Python FastAPI on the backend, powered by GPT-4o and advanced models.

Challenges we ran into

Common developer mistakes encountered during deployment, using your MP3 issue as an example- During deployment, we faced issues such as:

Missing or misreferenced files causing 404 errors (e.g., an MP3 missing from version.json broke audio playback).
Improperly handled HTTP errors like 400 Bad Request impacting API calls.
Incorrect asset paths leading to broken media or UI elements.
Incomplete or outdated resource updates causing inconsistent user experiences.

Tips to Avoid Common Deployment Issues

Verify all asset references: Double-check file paths and names to prevent 404 errors.
Implement robust error handling: Gracefully manage HTTP errors like 400 or 500 to improve resilience.
Use automated tests: Validate assets and API responses during build and deployment stages.
Keep resources updated: Ensure all files and data (like APIs or media) are current before release.
Monitor post-deployment: Track errors and user feedback quickly to catch missed issues.

Accomplishments that we're proud of

A working prototype that feels human and kind. It meaningfully helps anxious users (including thalassemia patients and families) by combining reflection tools with practical access to blood banks—bridging care and logistics.

What We Have Learned

Building AI for sensitive health contexts requires balancing technical precision with ethical responsibility.
Emotional support tools must prioritize privacy, consent, and empathetic communication.
Real-time NLP models need careful tuning to avoid overconfident or misleading responses.
Reliable, up-to-date data—especially for urgent resources like blood banks—is critical to user trust and safety.
Deployment demands rigorous asset management and thorough testing to prevent avoidable failures like missing files or broken links.

What's next for AI for Good hackathon project

MindMend/whisper is a caring AI helper that listens to users’ feelings through daily journaling and friendly chat. It supports thalassemia patients by quickly showing nearby blood banks when they need urgent help. The project combines emotional support with real-world resources to make a difference in people’s lives.