About the project
PiggyBank started as a side project with a clear vision: to give everyday service workers and performers a modern tool to receive appreciation in a cashless world. The project was developed end-to-end — from UX design and infrastructure planning to implementation and deployment — as part of a small remote team. I was responsible for the full stack development, DevOps setup, and technical architecture. The focus was on building something lightweight, intuitive, and scalable, while keeping onboarding friction low for users unfamiliar with tech tools.
What was the problem
In many industries where individuals perform or serve in physical spaces — such as street artists, bar staff, or hairdressers — earnings often rely on spontaneous, small-scale tipping. With the shift toward cashless societies, the absence of easy digital alternatives has created a significant gap: people are willing to give, but lack the means. For workers, this means losing income simply because modern payment tools aren't designed for real-world, informal transactions.
Project execution
To address this, we created a lightweight B2C platform that allows users to set up a digital profile that functions like a personal tip jar. Once registered, users receive a unique link that can be embedded into QR codes or written onto NFC stickers, which they can display in their work environment. When scanned or tapped by supporters, the system opens a mobile-optimized page where tips can be sent via digital payment — with no need for an app download or account creation on the sender’s side. This not only enables frictionless microtransactions in real-life scenarios but also builds visibility and legitimacy for independent workers.
Technical details
The application consists of a Nuxt frontend and a FastAPI backend. Nuxt was selected for its server-side rendering (SSR), Vue ecosystem, and SEO benefits, allowing every user profile to be instantly sharable and indexable. The backend, built in FastAPI, manages all core logic: account creation, profile editing, media uploads, and secure donation handling. We deployed on AWS, using S3 for image and asset storage, RDS for relational data management, and Lambda functions for QR/NFC generation logic. Authentication was implemented using JWTs with secure cookie fallback for client-side usage. The platform is Dockerized for scalability and CI/CD pipelines ensure fast, safe deployments. Stripe or PayPal integration (based on region) was used for donation flow, including fraud protection and user verification layers.