MediPal

Inspiration

Health is complicated. Jargon medical terms and complex prescriptions are difficult, especially for the millions of elderly users and non-native speakers who navigate health journeys with nothing but papers and zero oversight. Health is a lifelong conversation, not a series of isolated events. We built MediPal to be the 'digital family member' who sits in on every appointment, remembers every pill, and translates complex medical logic into a simple, supportive chat. Because when health is on the line, clarity is the best medicine.

What it does

MediPal makes your entire health journey as easy as having a chat. Our platform seamlessly helps patients before, during, and after medical appointments through accessible conversation. Prior to appointments, the web application integrates chat-based health check-ins with proactive, personalized prompting from the user’s digital caretaker. No generic “Ask Anything” prompt, rather MediPal will ask about previously logged symptoms or upcoming appointment concerns. Using the user’s personal data, chat history and calendar logs, a report for the patient can be downloaded to be used during doctor’s appointments ensuring a positive, productive experience in the office – doctors are getting detailed medical concerns and patients are getting accurate diagnoses. These reports include features such as medication adherence scores and timestamped symptom logs. After a medical appointment, users are guided to full recovery through medication reminders.

How we built it

The frontend of MediPal was built with React and hosted on vite.js We used Figma to create design mockups to bring our initial idea to life. On the backend, our goal wasn’t just to answer messages but to remember health information over time. When a user speaks or types, the frontend sends the message to our API. We use Dedalus Lab’s MCP to access Claude 4.5, which handles two things: natural conversation and reasoning over previous context. Important health details such as symptoms and medications taken are extracted and stored as structured records in MongoDB, not just chat logs. That lets us bring past information back into future conversations, and build a symptom timeline instead of isolated messages.

Challenges we ran into

This was our first time working with React and Expo, and early on we ran into issues just getting the mobile app to run reliably. With limited hackathon time, we had to make a decision: spend hours debugging setup or pivot. We chose to pivot to a web-based application, which let us move faster and focus on what actually mattered: the product logic, memory, and user experience. At the same time, we were learning multiple new technologies such as React, voice input, MongoDB, and Dedalus MCP, AI tools while building. The biggest learning was knowing when to simplify, adapt under pressure AND know to delegate tasks amongst each other for efficiency even given a small scope of this project.

Accomplishments that we're proud of

What we’re most proud of is that we actually shipped something people could use. We didn’t just build a chatbot, we built a system that remembers users, guides them gently, and keeps track of symptoms over time. We focused on accessibility from the start, especially for elderly users, large UI elements, voice-first interaction, and minimal friction. And despite this being our first time with many of these technologies, we made tough calls, pivoted when needed, and delivered a complete and functional experience.

What we learned

One of the key things we learned is how to leverage our sponsors’ tools, like Figma for rapid design and Dedalus for health data and AI reasoning. Using these tools not only helped us build MediPal faster, but also gave us insight into how powerful and convenient these platforms are which is something we can carry forward as we improve the product. It also made us think about the future growth and longevity of MediPal, and how we could continue iterating and scaling it after the hackathon.

What's next for MediPal

For the next steps, we want to expand MediPal into a mobile application with notifications, so users can get timely reminders for medications and symptom tracking. Security is also key so we plan to implement encrypted user authentication to protect sensitive health data. On the product side, we want to further enhance the conversational memory and voice interaction, and potentially support multiple languages to reach more users. Ultimately, we see MediPal as a personal health buddy that helps people stay on track, reduces stress after leaving the hospital, and makes health management simple for elderly users.