Team Dhurandhar

Inspiration Improving personal health

We were inspired by the abundance of health and fitness data generated by everyday devices—sleep trackers, fitness watches, and nutrition apps—but noticed that most of this data goes largely unused. Our goal was to build a system that could transform raw metrics into actionable, personalized guidance, helping people improve their wellness in a meaningful, measurable way.

What it does

Our solution is a personal Agentic AI health companion that integrates sleep, nutrition, and overall activity data to provide real-time insights and recommendations. It generates daily plans, automates routines such as workout scheduling and grocery lists, issues proactive alerts for metrics like HRV drops or sleep debt, and continuously learns from a user’s patterns to become smarter and more personalized over time.

How we built it

We leveraged a combination of workflow automation and AI orchestration tools, including n8n for workflow automation and AWS Bedrock for AI-driven insights. The system ingests data from multiple sources, processes it through AI models to generate actionable insights, and delivers personalized recommendations and automations to the user seamlessly.

Challenges we ran into

Multiagent interaction, setting up AWS bedrock, collecting data from fitness watch APIs

Accomplishments that we're proud of

We successfully built a working prototype that can analyze sleep, nutrition, and activity data and generate actionable recommendations. It is capable of automating routines and providing proactive alerts, demonstrating a fully functional personal health assistant. This marks a major step toward turning raw wellness data into meaningful, user-centered insights.

What we learned

We gained invaluable experience in integrating disparate data sources, building adaptive AI workflows, and translating complex health metrics into actionable recommendations.

What's next for Team Dhurandhar

Built With

• agentic
• ai