Bloom

The AI-powered financial ecosystem for the modern mom. Turn household clutter into cash and neighborhood connections into massive savings.

Comment

Inspiration

Across the globe, single mothers are the ultimate multitaskers, yet they are often overlooked by traditional fintech. Whether in Johannesburg, London, or New York, the challenge is the same: how to stretch a budget when time is your most scarce resource. The inspiration for Bloom came from seeing friends in my community prepare for motherhood. I realized they didn't need another "budget tracker"—they needed a financial engine that turned their limited time into real income and their neighborhoods into a support system.

What it does

Bloom is a social-commerce ecosystem designed to help parents Save and Earn regardless of their location:

The AI Side-Hustle: Using a reasoning-vision pipeline, users can scan any item at a retail store to instantly see its market resale value. Bloom turns a 10-second photo into a professional marketplace listing, making "side-hustling" possible for the busiest parents.

Village Grids (Bulk-Buying): Bloom brings the "Village" back by allowing neighbors to split bulk purchases. By grouping orders, users unlock wholesale prices on essentials like diapers and formula, saving up to 30% on the costs of raising a family.

Secure Chat: A real-time hub for coordination, neighborhood safety, and community support.

How we built it

I built Bloom to be as high-performance as the parents who use it:

Frontend: Built with React Native for a seamless, "Power Pink" cross-platform experience.

Backend: A C# .net 9 service deployed on Render, utilizing Socket.io for real-time community updates.

Database: MongoDB Atlas handles our geospatial data, allowing us to map "The Village" within a 5km radius of any user globally.

Intelligence: I integrated Phi-4-multimodal for visual product recognition and DeepSeek-R1 for complex market reasoning and pricing logic.

Revenue: RevenueCat manages our "MamaPro" subscriptions, allowing the app to scale globally with ease.

Challenges we ran into

The biggest challenge was the "Real-World Test." Retail environments vary globally, with different lighting, price tag formats, and data speeds. I had to optimize the AI pipeline to handle varied image qualities and implement a robust caching strategy so the app feels fast, even in areas with spotty connectivity. Bridging the latency between visual identification and financial reasoning while maintaining a responsive UI was a significant technical hurdle.

Accomplishments that we're proud of

I am incredibly proud of getting the DeepSeek-R1 reasoning to work in real-time on a mobile device. Seeing the AI actually "think" through the resale value of a product while standing in a busy aisle was a huge technical win. Successfully integrating MongoDB Change Streams for the real-time chat also felt like a major milestone in making the app feel like a living, breathing community.

What we learned

I learned that the best technology is invisible. A busy parent doesn't care about "Large Language Models"; they care about the bottom line. Building Bloom taught me to bridge the gap between complex engineering and human necessity—focusing on impact over jargon.

What's next for Bloom

The next step is scaling the "Village" feature to support international retail partnerships. I want to allow users to pick up their "Bulk-Buy" orders directly from major global retailers using shared QR codes. I also plan to implement an AI-driven "Trust Score" to make the Bloom marketplace the safest global environment for parents to connect, trade, and thrive.

Built With

Share this project:

Updates