​
About

I, Surya Deep Singh and Suyash Gupte — had an amazing 30-hour sprint at the AI Hackathon with Meta Llama, building a solution under the theme “AI for Societal Good” using Meta’s Llama foundation models.

​
Overview

In the rapidly evolving EdTech landscape, educators face challenges in creating high-quality assessments at scale, while students often receive generic, one-size-fits-all evaluations that fail to adapt to their individual learning needs. There is a need for an intelligent system that can automate question generation, personalize assessments in real-time, and deliver actionable feedback — all while reducing the burden on teachers and enhancing student engagement and outcomes.

​
Architecture

architecture

​
Solution

We developed a Generative AI-powered Adaptive Assessment System aimed at transforming the EdTech space by:
  • Automating Question Generation: Helping teachers save time with AI-generated questions.
  • Adaptive Learning Engine: Dynamically adjusting question difficulty in real-time based on each student’s performance.
  • Personalized Feedback: Offering students instant, tailored feedback to accelerate learning.
Our system supports teachers and empowers students, showcasing the potential of GenAI in delivering equitable and scalable education.

​
Skills picked up

  • LangChain, LangGraph
  • Qdrant Vectorstore
  • Amazon Bedrock
  • Developing Responsible AI Agents