Inspiration
What it does
How we built it
Challenges we ran into
Accomplishmen## Inspiration
🔥 What Inspired Us In today’s fast-paced world, millions delay seeking medical help due to lack of access, fear, or uncertainty—especially in rural or underserved areas. We were inspired by a single, powerful idea: What if anyone, anywhere, could ask a virtual doctor for help—instantly? This led us to create HealthMate, a personal health companion that offers real-time guidance and empowers users to take control of their well-being.
📚 What We Learned Throughout this journey, we deepened our understanding of Large Language Model (LLM) integration, particularly using LLaMA 3 (70B) via the Groq API, and explored LangChain to manage prompt flows effectively. We learned how to build secure, session-based applications in Streamlit, how to incorporate geolocation tools like Folium & Geopy, and how to generate downloadable documents using python-docx—all while designing a seamless and interactive user experience.
🛠️ How We Built It HealthMate was built using Streamlit as the front-end framework with Python for backend logic. We integrated Groq’s LLM API through LangChain’s ChatGroq, enabling dynamic symptom diagnosis. We used Geopy + Folium to find nearby doctors based on user location, built a lightweight reminder engine using streamlit_autorefresh, and created a Word-based chat history export with python-docx. The entire UI was carefully designed to be intuitive, responsive, and informative.
⚠️ Challenges We Faced 🔄 Designing LLM prompts to give meaningful, medically relevant answers without hallucination
⏰ Building a reminder engine in Streamlit without persistent background services
🧠 Balancing AI flexibility with user trust, especially in healthcare
🗃️ Operating without a backend database, relying solely on session state
Despite these challenges, we stayed focused on usability, privacy, and user empowerment—making HealthMate a strong contender in digital healthcare innovation. ts that we're proud of
Prototype -Username:admin and Password:1234
Built With
- folium
- geopy
- langchain
- llama
- python
- python-docx
- streamlit
- streamlit-autorefresh
Log in or sign up for Devpost to join the conversation.