GenderAI
From AI Experiment to Full-Stack SaaS Product
I architected, trained, and deployed a custom neural network as a full-stack SaaS application. The final product features a public API, a web tool for bulk CSV processing, and a token-based payment system that generates revenue from paying customers.
Role | Sole AI & Full-Stack Developer |
Timeline | 3 Weeks Over Christmas |
Tech Stack | Angular, Node.js, Express, synaptic, Stripe |
The Challenge:
Productizing a Real-World AI Application
As an engineer, I believe in learning by doing. I wanted to move beyond pure theory and master the practical challenges of the entire AI product lifecycle. My goal was not just to build a model, but to turn it into a complete, self-service SaaS product with real, paying customers.
The challenge I set for myself was to:
- Source and clean a large dataset suitable for training.
- Train and deploy a custom neural network.
- Build a full-stack application with a user-friendly frontend and a public API.
- Successfully monetize the service with a token-based payment system.
My Approach:
Engineering a Full-Stack AI Application
This project was an exercise in practical, end-to-end product development.
The Full-Stack Architecture
I chose a comprehensive full-stack JavaScript approach for the application layer. The backend was built with Node.js and Express to serve as a robust API interface. For the frontend, I used Angular to create a dynamic and responsive web experience.
From Model to User-Facing Tool
The application was designed to be a complete, self-service tool. I built a web interface that allows users to get results directly in their browser—either by entering a single name or, for more advanced use cases, by uploading entire CSV lists for bulk processing.
Token-Based Monetization
To turn the tool into a viable product, I designed and integrated a flexible, token-based payment system using Stripe. Users could purchase the exact number of tokens they needed, which were then consumed with each name processed. This model provides a clear and scalable path to revenue for the API and web tool, and it reduces the hesitation to pay too much, which has a positive effect on conversion.
The Result:
A Profitable AI SaaS Product
The final result was a fully functional, revenue-generating SaaS application. By building a user-friendly tool and integrating a seamless, token-based payment system, the project successfully attracted paying customers and generated approximately 500 CHF in revenue. This proves not just technical feasibility but also successful market validation.
This project showcases my ability to:
- Manage the end-to-end lifecycle of an AI product—from data to deployment to revenue.
- Build a complete full-stack application with Node.js, Express, and Angular.
- Design and integrate a flexible, token-based payment system with Stripe.
- Bridge the gap between data science and practical, commercial application development.
Let’s Build Intelligent Applications
If you’re looking for a partner who can not only build your application but also integrate intelligent, data-driven features with a clear path to monetization, I’d be happy to talk.