AI Chinese — AI Course Generator for an Online School
AI platform for automatic generation of Chinese language courses based on the proprietary methodology of online school China Campus Network.
Challenge
The Client's Challenge
Online school China Campus Network created courses with a team of 7 methodologists. Developing one short course took a month. This became a growth bottleneck: scaling the product line at that pace was impossible.
The project was initiated by the CEO: the company needed both an innovative AI product to support Skolkovo residency applications and a real tool to optimize internal processes. The service users are methodologists who previously built everything by hand.
Solution
What We Built
We built a web service with a methodologist's personal account where course generation starts by configuring 15+ parameters. The school's proven methodology — module, lesson, and exercise structure — is baked into the system. Under the hood: a mix of LLMs (OpenAI + Chinese neural networks) for the optimal balance of quality and inference cost.
Key Features
- Configure target audience, topics, and difficulty level
- Upload additional materials to enrich the course
- 15+ customization parameters for structure and content
- Built-in school methodology: modules → units → lessons → tasks
- LLM mix (OpenAI + Chinese neural networks) for cost optimization
- Edit and refine generated content
Technologies
Tech Stack
Results
What the Client Achieved
Галерея
Project Screenshots
“We weren't sure it was even feasible — we had no AI expertise ourselves. From the very first meetings, Frog Studios showed what was possible and how to get there, offering multiple solution options. Even when we hit dead ends, a way forward always emerged that we were fully satisfied with. We got exactly the product we wanted and continue working together on support and development.”
Mikhail
CEO, China Campus Network
Other Cases
Related Projects
Start a project
Ready to discuss the project?
Leave a request — we'll respond within an hour. At the first meeting we'll break down the task, show relevant cases and suggest the right entry package.