MedBot — Your AI-Powered Medication Assistant
The idea for MedBot was born from a deeply personal place. Our team member's mother, a CEO of a multinational company, juggles an incredibly demanding schedule every day — and despite being diabetic, she often forgets to take her medication simply because life gets in the way. At the same time, their 80-year-old grandfather struggles to remember which medications to take and when, making it difficult for him to manage his health independently. We realized this isn't just their family's problem. Millions of busy professionals, elderly individuals, and caregivers face the same challenge every single day — managing complex medication schedules without any real support system. According to studies, medication non-adherence causes nearly 125,000 deaths annually in the US alone. We built MedBot to change that.
MedBot is an AI-powered medication assistant that makes managing prescriptions effortless for everyone — from busy executives to elderly patients.
Here's what MedBot can do:
- Smart Prescription Reading — Simply upload your prescription in any format (PDF, Word, Excel, or text) and MedBot automatically reads and understands it using AI
- Voice Reminders — MedBot announces medication reminders out loud at the right times, one by one, so nothing gets missed
- Ask Anything — Patients can ask questions about their medications, dosage, allergies, or next doctor's appointment in plain conversational language — by typing or speaking
- Live Voice Conversation — Talk to MedBot in real time like a conversation and get spoken answers back instantly
- Manual Entry — No prescription document? No problem — patients or caregivers can manually enter medication details
- Personalized Experience — Every user gets their own personalized reminder schedule based on their specific prescription and preferred times
MedBot is built on a modern AI-powered backend using Python and Flask. Here's our tech stack:
Backend: Python, Flask AI & NLP: Anthropic Claude API for prescription parsing and intelligent Q&A Voice: OpenAI Whisper for speech to text, gTTS for text to speech Document Processing: pdfplumber, python-docx, openpyxl for multi-format prescription reading Scheduling: APScheduler for intelligent medication reminder scheduling Frontend: HTML, CSS, JavaScript
Challenges We Ran Into
- Getting the AI to reliably extract structured medication data from unstructured prescription documents in various formats
- Building a real time voice pipeline that listens, understands, and responds naturally without cutting off the user
- Designing a smart scheduling system that works both with specific times set by the patient and automatic time calculation based on prescription frequency
Accomplishments That We're Proud Of
- Built a fully functional AI powered prescription reader that works with any document format
- Created a seamless voice conversation experience for patients who struggle with typing
- Designed a reminder system that announces medications one by one with a natural voice — making it genuinely accessible for elderly users
- Built the entire backend from scratch in under 24 hours
What We Learned
- How to build and chain AI pipelines together for real world use cases
- How to process and extract meaningful structured data from unstructured medical documents
- The importance of accessibility in health tech — designing not just for tech savvy users but for elderly and busy users who need simplicity above everything
What's Next for MedBot
- Mobile app with push notifications for on the go reminders
- Caregiver mode — family members can monitor and manage medications for elderly loved ones
- Multi-language support — so MedBot can serve non-English speaking patients
- Doctor integration — direct prescription import from doctor's office systems
- Medication interaction warnings — AI powered alerts when medications could interact dangerously
Built With
- anthropic
- apscheduler
- css
- flask
- gtts
- html
- javascript
- openai
- pdfplumber
- python
- typescript
- vscode
Log in or sign up for Devpost to join the conversation.