Camge.org

What it does

The app is based on the theory of personality from Florence Littauer's book, "Personality PLUS." We've refined the questions and answers to ensure a more accurate and engaging experience for users. The app asks users 20 yes or no questions, and then uses the Gemini Large Language Model (LLM) to generate a personality description based on their answers.
In the Gemini competition, we leveraged the knowledge base from the personality test to create a web application that generates 20 yes or no questions. Users answer the questions, and then send their answers to the Camge app, which uses the Gemini LLM to generate a personalized personality description.
From a technical perspective, the app is built using Angular for the frontend and Java Spring Boot for the backend. The frontend is divided into two components: "Question-ai" and "Answer-ai." The "Question-ai" component calls the "getSurveyQuestions" service, which generates the questions using Gemini. The "Answer-ai" component prints the final personality evaluation and recommendations for human improvement.
The backend is built using Java Spring Boot, and it calls Gemini's REST API using the "gemini-pro" model. Given the knowledge base from the "Personality PLUS"; Gemini generates 20 yes or no questions. In the second iteration, we send the user's answers to Gemini, and it helps us generate a personality description.