AI-Driven Test Case Generation: My Journey from Code to Cloud

Tech Stack: Cursor (AI) + React + Netlify Functions + Gemini/Ollama + Cloudflare

TL;DR: Key Takeaways

• Open-Source Demo: Built AI test case generator as a personal research project from concept to deployment
• Hybrid AI Architecture: Gemini (cloud) + Ollama (local) for flexible AI processing
• Serverless Infrastructure: Netlify Functions + Cloudflare for scalable deployment
• Architecture Deep Dive: Complete walkthrough of system design and challenges
• Open Source: Full codebase available on GitHub for engineering teams

---

The Challenge: Designing Tests, Faster

As a Quality Engineering leader, I know the pain: crafting robust test cases from dense requirement documents is crucial, yet time-consuming. My mission was to build AI Test Case Generator, an app that turns requirements into structured test suites instantly.

The journey wasn't just about the AI; it was about building a resilient, scalable platform from the ground up, tackling deployment complexities along the way.

---

The Architecture: From Conflict to Clarity

Initially, managing my professional portfolio alongside this powerful AI tool led to "nested repo" deployment headaches. My solution was a strategic migration to Netlify, simplifying routing and leveraging its powerful serverless capabilities.

Here's the breakdown of the tech that made it happen:

1. AI-Powered Development: Cursor

My development partner for this entire project was Cursor, an AI-native IDE. It allowed me to rapidly refactor code, streamline deployment processes, and focus on the core AI logic rather than getting bogged down in boilerplate or debugging configuration conflicts. Cursor was key to accelerating development and maintaining a lean codebase.

2. Flexible AI Engines: Gemini 1.5 Flash & Ollama

For the intelligence layer, I wanted flexibility:

• Google Gemini 1.5 Flash: Used for its immense context window and speed, ideal for processing large PDF requirements documents and generating comprehensive test cases quickly in the cloud.
• Ollama (Local AI): For developers and testers who prefer privacy or local control, I designed the backend to seamlessly integrate with Ollama. This allows users to leverage powerful open-source LLMs running directly on their machine, providing an alternative for offline or sensitive data processing.

3. Serverless Backend: Netlify Functions

The brain of the operation runs on Netlify Functions. These lightweight, serverless APIs handle:

• PDF Parsing: Extracting text from uploaded documents.
• AI Orchestration: Communicating with both Gemini and Ollama.
• Security: Implementing IP-based Rate Limiting directly in the function code to protect against abuse and manage API costs.

4. Global Reach & Security: Cloudflare

My domain and overall security posture are managed by Cloudflare. It provides robust DNS, SSL, and network edge caching, ensuring the application is fast, secure, and globally accessible.

5. Frontend: React & Tailwind CSS

The user interface is built with React and styled with Tailwind CSS, delivering a responsive, intuitive experience for quality engineers.

---

The Journey Visualized

The architecture follows a clean, serverless flow. Here's how it works:

All wrapped in Cloudflare for global distribution and security.

---

Why This Matters for QE Leaders

This project is more than just an app; it's a blueprint for modern quality engineering. It demonstrates how a strategic blend of AI development tools, flexible AI models, and robust serverless architecture can transform how we approach test design, driving efficiency and innovation.

Experience it yourself: testcase-ai.suneetmalhotra.com

---

Open Source & Community

I'm a firm believer in the "Quality First" community. I've made the entire codebase for this generator—including the Netlify Function logic and the Gemini/Ollama orchestration—available on GitHub. Feel free to fork it, contribute, or use it as a template for your own AI projects.

GitHub Repository: SuneetMalhotra/ai-test-case-generator

---

About the Author: Suneet Malhotra is an AI-Driven Quality Engineering Leader with 20+ years of experience in scaling complex platforms.