The Business Challenge
The business situation the organisation faced before the project began.
Instructional designers were spending an average of 6 hours manually researching and structuring course curriculums. Furthermore, integrating LLMs directly into the frontend posed security risks (API key exposure), and there was no mechanism to accurately track and control the LLM API budget per user session.
How the Solution Was Delivered
Discovery, design, development, testing and rollout — the journey, not the tooling.
- 1Architecture Design & Tech Stack Selection (FastAPI, Next.js, SQLite)
- 2Backend Development: API endpoints, JWT Auth, and Gemini LLM integration
- 3Database Engineering: Schema design for conversational memory and token tracking
- 4Frontend Development: Building the responsive UI and connecting APIs
- 5DevOps & Deployment: Linux VPS setup, Nginx reverse proxy, and PM2 process management
Business Outcomes & Impact
The measurable outcomes recorded on this AI Case Study — with the story behind them.
Reduced manual curriculum design time by 99.4% (from 6 hours to under 2 minutes) and achieved 100% API cost visibility.
Business Outcome Categories
The story behind the numbers
The system reduced the manual curriculum planning process from 6 hours to under 2 minutes per course. It established a secure, production-ready infrastructure with 100% visibility into API costs, preventing budget overruns and ensuring high availability.
Lessons Learned
What surprised the team, what worked well, and what would be done differently.
- Decoupling the frontend from the AI processing layer via a FastAPI backend is crucial for security and CORS management.
- Building a custom database for token tracking is essential for sustainable AI scalability in enterprise environments.
- Using a reverse proxy (Nginx) and process manager (PM2) is non-negotiable for zero-downtime AI deployments.