Fetebird revolutionizes assignment grading by harnessing the power of the Gemini API(gemini-1.5-flash-001) and a Retrieval Augmented Generation (RAG) architecture. It streamlines the evaluation process, providing educators with valuable insights into student understanding while maintaining flexibility and adaptability. Core Functionality Reference Answer Database - At its heart, Fetebird maintains a database of verified reference answers, each vectorized for efficient comparison. Initially, this database may contain only the teacher's solution, but it expands dynamically as more student submissions are reviewed and approved. Gemini API Integration - The app leverages the Gemini API (gemini-1.5-flash-001) and the Langchain4j Java library to interact with a large language model. This enables Fetebird to perform complex evaluations and provide detailed feedback. RAG Architecture - When a professor initiates an AI evaluation, Fetebird employs a RAG architecture. It retrieves relevant reference answers from the database and presents them to the Gemini model alongside the student's submission. Prompt Engineering: Carefully crafted system and user prompts guide the model's evaluation. It considers the reference answers and the student's work to provide comprehensive feedback, highlighting strengths, weaknesses, and areas for improvement.
Built with
Web/Chrome
Team
By
Fetebird
From
Australia
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Missing the information I need","missingTheInformationINeed","thumb-down"],["Too complicated / too many steps","tooComplicatedTooManySteps","thumb-down"],["Out of date","outOfDate","thumb-down"],["Samples / code issue","samplesCodeIssue","thumb-down"],["Other","otherDown","thumb-down"]],[],[],[],null,["# Fetebird\n\n[More Apps](/competition/vote) \n\nFetebird\n========\n\nUltimate solution for University \nVote \nVoted!\nWhat it does\n\nFetebird revolutionizes assignment grading by harnessing the power of the Gemini API(gemini-1.5-flash-001) and a Retrieval Augmented Generation (RAG) architecture. It streamlines the evaluation process, providing educators with valuable insights into student understanding while maintaining flexibility and adaptability. \nCore Functionality \nReference Answer Database - At its heart, Fetebird maintains a database of verified reference answers, each vectorized for efficient comparison. Initially, this database may contain only the teacher's solution, but it expands dynamically as more student submissions are reviewed and approved. \nGemini API Integration - The app leverages the Gemini API (gemini-1.5-flash-001) and the Langchain4j Java library to interact with a large language model. This enables Fetebird to perform complex evaluations and provide detailed feedback. \nRAG Architecture - When a professor initiates an AI evaluation, Fetebird employs a RAG architecture. It retrieves relevant reference answers from the database and presents them to the Gemini model alongside the student's submission. \nPrompt Engineering: Carefully crafted system and user prompts guide the model's evaluation. It considers the reference answers and the student's work to provide comprehensive feedback, highlighting strengths, weaknesses, and areas for improvement. \nBuilt with\n\n- Web/Chrome \nTeam \nBy\n\nFetebird \nFrom\n\nAustralia \n[](/competition/vote)"]]
