BumpBuddy

Inspiration

The inspiration for building BumpBuddy came from my own experience as a first-time mom. During pregnancy, I realized how overwhelming it can be to navigate the huge amount of information available online. Every stage of pregnancy brings new questions about baby development, nutrition, health, and emotional well-being, yet it can be difficult to find guidance that truly fits your personal situation such as your pregnancy stage, dietary preferences, or health conditions. Even though family and friends may want to help, they often cannot fully understand what a mother is experiencing physically and emotionally.

Through this journey, I realized that many mothers face the same feelings of uncertainty and isolation. This experience also made me more aware of a larger issue in maternal health. A major review of maternal deaths in the United States found that about 84% of pregnancy-related deaths were preventable with proper care, education, and timely intervention (Source: https://www.aimcci.org/the-state-of-maternal-health/). This highlights how critical it is for mothers to have access to reliable information, early warning signs, and personalized support during pregnancy.

That realization inspired me to create BumpBuddy, an AI-powered companion designed to provide personalized week-by-week guidance and instant answers to pregnancy-related questions—helping mothers feel more supported, informed, and confident throughout pregnancy and early motherhood.

What it does

• Generates personalized weekly guidance based on pregnancy week, age, dietary preferences, health conditions, and activity level
• Provides comprehensive information covering baby development, mom's body changes, nutrition advice, safe exercises, warning signs, mental wellness tips, and weekly checklists
• Offers an interactive chat assistant where users can ask any pregnancy-related questions and receive instant, context-aware answers
• Includes medical disclaimers to ensure users understand this is educational content and should consult healthcare providers for medical decisions

How we built it

We built front end with HTML/CSS/JavaScript application hosted on AWS S3 with static website hosting. For the backend, we used serverless architecture using AWS Lambda (Python 3.14) for compute. For AI Models, we used Amazon Bedrock with Nova 2 Lite for both guidance generation and chat responses. The API Layer is Amazon API Gateway (REST API) with CORS enabled for cross-origin requests. We have Amazon DynamoDB for storing user profiles and guidance history.

Challenges we ran into

We had challenges to build the architect and learnt about new technology in a short amount of time. We spent time trying to design the app so that it is easier for usage and also works well using Nova 2 Lite.

Accomplishments that we're proud of

We were able to build a fully functional AI health assistant in a short timeframe using cutting-edge AWS services. We implemented comprehensive pregnancy guidance covering 8+ critical topics that first-time mothers need and created a seamless user experience with both structured guidance and conversational chat. We also leveraged Amazon Nova 2 Lite effectively for high-quality, empathetic AI responses tailored to pregnancy health

What we learned

• Amazon Bedrock is for domain-specific AI applications with proper prompt engineering
• Nova 2 Lite for generating output and chat answers
• Serverless architecture significantly reduces infrastructure complexity and costs
• CORS is critical for web applications
• API Gateway deployment is a separate step from creating resources

What's next for Pregnancy Assistant

We want to implement chat history storage in DynamoDB and create weekly email notifications using Amazon SES. I also want to add meal planner feature with recipe suggestions and hospital preparation checklist for third trimester.

Built With

• amazon-bedrock
• aws-lambda
• css
• dynamodb
• html
• javascript
• nova-2-lite
• python
• s3