Inspiration
This is a real-life situation I'm facing. Aidan, our third child, is extremely talented in archery. At 11 years old, he has competed twice (since he was 9) at the national and international levels. He has won numerous championships, mostly in higher age groups. Aidan has now graduated from elementary school. To maintain his performance and improve his skills, he needs more time to practice. Unfortunately, at school, his time is spent at school, leaving him with only one or two hours in the afternoon. This is certainly not enough. We still prioritize Aidan's education, but we also want to ensure that he continues to develop in archery. Therefore, we are planning to homeschool Aidan.
What it does
- Identify your child's personality and learning style through assessment
- Build a personalized weekly curriculum
- Present daily material tailored to your child's age and interests
- Serve as an AI tutor that can be engaged in discussions at any time
- Help parents monitor learning progress in real-time
How we built it
The prompts are divided into several parts:
- Prompts for AI personalization. These form the AI's personality.
- Prompts for child personality. These are generated by the AI when analyzing the assessment results. These assessment results are saved and used as child personality prompts.
- Prompts specific to each feature
This way, the system will send prompts with the formula: AI Personality Prompt + Child Personality Prompt + Function Prompt.
These prompts are tested in Google AI Studio while selecting the best model. Gemini is then used to build the application, including deployment to GCP.
Challenges we ran into
Alignment with government education standards.
Accomplishments we're proud of
The system can now perform its core functions and utilize Gemini as the backbone of the critical process: creating a personalized curriculum, interactive Q&A, and assessing children's personalities.
What we learned
Utilizing Google AI Studio and Gemini opens up opportunities and makes it easier to build applications that address real-world problems.
What's next for AI and: AI for Homeschooling
- Adding interactive visualization features to learning by utilizing Nano Banana.
- The ability to adjust curriculum materials based on daily evaluations, allowing AI to proactively and adaptively develop curriculum materials to continuously improve children's learning performance.
- More active parent involvement by utilizing Gemini's suggestions.
Log in or sign up for Devpost to join the conversation.