• 

  Frontend UI result

• 

  Frontend code

• 

  Backend code

• 

  Final look

CBC Smart Study Assistant

Inspiration

The Competency-Based Curriculum (CBC) in Kenya encourages creativity, practical learning, and personalized instruction.
However, many teachers spend hours preparing lesson notes, quizzes, and real-life activities manually.
As a developer passionate about both AI and education, I wanted to build a tool that saves teachers time, supports their creativity, and makes learning engaging for students.

That’s how the CBC Smart Study Assistant was born an AI-powered solution that instantly creates complete CBC lesson plans.

What it does

The CBC Smart Study Assistant uses Google Gemini AI to generate:

• A short and clear lesson summary
• Three multiple-choice quiz questions
• One real-life classroom activity
• saving and exporting generated lesson plans as PDFs.

Teachers or students simply type a topic like "Photosynthesis in plants", and the app instantly produces structured, easy-to-understand CBC learning content ready to teach or study.

How we built it

• Frontend: Built with Next.js and styled using TailwindCSS for a clean, responsive, and minimal UI.
  
• Backend: Developed with Node.js and Express to handle API requests and responses.
  
• AI Engine: Integrated Google Gemini 2.5 Flash API to generate accurate, structured educational content.
  
• Testing: Verified backend routes using Postman before connecting to the frontend.
  

The backend receives a topic, sends it as a prompt to Gemini, and returns a formatted response containing a summary, quiz, and activity — all tailored to CBC standards.

Challenges we ran into

1. Gemini model errors: Initially, the model gemini-1.5-flash returned 404 Not Found. I resolved this by switching to the supported model gemini-2.5-flash.
   
2. Tailwind setup issues: A PostCSS error occurred (tailwindcss used directly as a PostCSS plugin). Solved by installing @tailwindcss/postcss and reconfiguring PostCSS.
   
3. Language tuning: Early outputs included Swahili phrases. I refined the prompt to enforce English-only responses.

Accomplishments that we're proud of

• Built a fully functional AI-powered educational tool in less than a day.
  
• Seamlessly integrated Gemini API with a clean and interactive Next.js frontend.
  
• Created lesson plans that are aligned to the CBC structure — summary, quiz, and activity.
  
• Showcased how AI can transform education in Kenya by improving efficiency and creativity.

What we learned

• How to effectively integrate and prompt-tune Gemini API for domain-specific results.
  
• The importance of structured prompts and clear formatting in educational AI applications.
  
• Practical debugging of Tailwind/PostCSS setups in a modern Next.js environment.
  
• That AI can truly complement human teaching — not replace it.

What's next for CBC Smart Study Assistant

• Add user authentication (for teachers & students).
  
• Introduce voice-based queries for young learners.
  
• Expand to support Grades 1–8 and all CBC subjects.
  
• Train Gemini further using local CBC datasets for even more accurate outputs.

“AI for Smarter Learning — Empowering Kenya’s Education Future.”

Built With

• express.js
• googlegeminiapi
• next.js
• node.js
• tailwindcss