CoinCrew

Inspiration

Many kids grow up without being taught how money actually works. This problem is even more serious for foster youth, who often lack consistent adult guidance when it comes to finances, government paperwork, and long-term planning. We noticed that most financial tools are built for adults and are overwhelming even for them. CoinCrew was inspired by the need to give kids and teens a safe, approachable way to learn financial literacy early and build confidence around money.

What it does

CoinCrew is a financial literacy platform designed for kids and teens, with a special focus on foster youth. The app teaches money management through interactive quizzes, spending simulations, and rewards. Users earn coins by completing quizzes and learning modules, then practice smart spending by using those coins on virtual items like clothes, food, toys, and accessories.

CoinCrew also includes a document analysis feature that allows users to upload complex financial or government documents such as benefit letters, tax notices, banking statements, or program eligibility forms. These documents are simplified into clear, age-appropriate explanations so users understand what the document means and what steps they should take next. An in-app chat assistant allows users to ask follow-up questions while viewing their documents.

How we built it

CoinCrew was built using TypeScript and Next.js (App Router) to create a structured, scalable web application with strong type safety. We used Next.js Route Handlers (/app/api/.../route.ts) to define a clean JSON-based API for features like quizzes, document processing, and chat interactions.

For data storage, we used MongoDB Atlas with Mongoose to manage user profiles, quiz progress, coin balances, and document metadata. To prioritize privacy, we do not store raw government document text. Instead, only non-sensitive metadata is saved, ensuring user information remains protected.

AI-powered explanations and document simplification are handled using the Gemini API, which allows us to transform complex financial and government language into age-appropriate, easy-to-understand summaries. This same AI system supports follow-up questions through the in-app chat feature.

The frontend was styled using Tailwind CSS, allowing us to quickly iterate on a responsive and accessible design that works well for both kids and teens without feeling childish. For deployment and demos, we used Vercel, enabling fast iteration and seamless hosting of our Next.js application.

Challenges we ran into

One of the biggest challenges we faced was merging multiple branches into a single cohesive codebase. Since our team members were working in different environments and on different features in parallel, we frequently encountered merge conflicts, dependency mismatches, and inconsistent file structures. Resolving these required careful coordination, clear communication, and a deeper understanding of Git workflows to avoid breaking existing functionality.

Another major challenge was learning how to effectively use the Gemini API within the limited timeframe. We had to experiment with prompt design, response formatting, and error handling to reliably generate simplified explanations and meaningful follow-up questions. Understanding how to guide the model to produce age-appropriate, accurate, and safe responses, especially for sensitive financial and government content.

Furthermore, another challenge was handling document uploads and rendering them reliably in the browser while keeping the experience simple for younger users. We also had to carefully manage client-side state across pages without causing hydration or rendering issues.

Designing for kids and teens was another challenge. We had to balance clarity and professionalism while ensuring the interface stayed engaging and easy to understand. Making financial concepts feel approachable without oversimplifying them required constant iteration.

Accomplishments that we're proud of

We’re proud of building a platform that tackles financial literacy in a practical and empathetic way, especially for foster youth who often lack access to consistent financial guidance. The document analysis feature stands out as a tool that makes intimidating government and financial paperwork understandable for young users. We’re also proud of creating an experience that teaches smart spending habits through interaction rather than lectures, while aiming to make a positive social impact by empowering vulnerable youth. Delivering a polished, cohesive frontend within a short time frame further reflects our team’s collaboration and execution.

What we learned

We learned that building for younger audiences requires thoughtful design choices and strong attention to usability. Technically, we gained deeper experience with client-side rendering, state management, and file handling in modern web applications. More importantly, we learned how impactful technology can be when it’s built with empathy and real-world problems in mind.

We gained hands-on experience working with the Gemini API, MongoDB Atlas, and TypeScript, while becoming more comfortable with React, Next.js, and Tailwind CSS in a real-world application. Beyond the tech, we learned how impactful software can be when it’s built with empathy and focused on solving meaningful, real-life problems.

What's next for CoinCrew

Next, we plan to expand our existing backend to support deeper document analysis and more personalized explanations for users. We also want to grow the quiz system with real-world budgeting scenarios, introduce long-term progress tracking, and refine rewards to better reinforce smart financial decisions. Long term, we hope to partner with foster care organizations and schools to make CoinCrew accessible to the kids who need it most and scale its positive social impact.