Inspiration
One of my relatives, a healthcare professional, is spending several hours after hospital hours completing documentation required for insurance providers. This manual process is time-consuming, mentally exhausting, and takes away from valuable personal time, leading to burnout and reduced efficiency.
What it does
RxNote.ai is your all-in-one AI companion for healthcare professionals — streamlining medical documentation, assisting with accurate coding, and accelerating clinical research. From helping physicians capture key insights to auto-suggesting the right billing codes and surfacing relevant research, RxNote.ai reduces admin overload and boosts productivity. It’s not just smarter notes — it’s intelligent support across your entire clinical workflow.
How we built it
We leveraged Deepgram for real-time, high-accuracy audio transcription of clinical conversations. The transcribed text is then passed through a finely-tuned LLM (Large Language Model) using carefully crafted prompts to generate structured and context-aware clinical notes. For medical coding, we integrated the Perplexity API to intelligently search and map appropriate billing codes based on the conversation context — ensuring accuracy and compliance with insurance documentation standards.
Challenges we ran into
One of the biggest challenges was ensuring accurate transcription for non-native English speakers. Accents, medical jargon, and varying speech patterns made it difficult to consistently capture conversations with high precision. This pushed me to fine-tune the transcription pipeline and explore ways to improve language understanding for more inclusive and reliable results.
Accomplishments that we're proud of
After building the MVP, I shared it with my relative, a practicing healthcare professional. He started using it during his clinical routine and was genuinely impressed with the accuracy and quality of the generated notes. His positive experience confirmed the real-world value of RxNote.ai and motivated me to keep pushing the product forward.
What we learned
Through this project, I learned how to build an AI-powered healthcare app from end to end — integrating transcription, language models, and external APIs to solve real-world problems. My goal is to continue improving RxNote.ai and make it accessible to healthcare practitioners at a minimal cost. You shouldn’t have to rely on expensive enterprise software to get intelligent, reliable support in your daily clinical workflow.
What's next for AI copilot for Healthcare professionals
I’m committed to evolving this into a full-fledged SaaS product, continuously improving it based on real feedback from healthcare professionals. I’ll be actively reaching out to users to understand their unique pain points and refine the solution accordingly. The long-term vision is to grow this into a startup company and build a powerful AI agent — like a virtual healthcare employee — that supports clinicians across documentation, research, coding, interpretation, and beyond. The future of healthcare needs smarter tools, and I’m excited to help lead that transformation.
Built With
- amazon-web-services
- clerk
- deepgram
- firebase
- javascript
- neon
- netlify
- nextjs
- openai
- perplexity
- postgress


Log in or sign up for Devpost to join the conversation.