Children learning

The Equity Learning Collective

A nonprofit organization created an AI-powered tool designed to create custom lesson plans for students by analyzing their historical performance, enhancing the learning experience.

Industry

Nonprofit

Teams & Services

Tech & Tools

Amazon Bedrock, Lambda, DynamoDB, S3, OpenSearch, EventBridge, Serverless

Key Data Points

This nonprofit organization had the idea for an AI-powered lesson planner that uses historical student performance data in order to define what exercises & external course content each student should consume.
They partnered with Protagona to build out this idea as POC to use for additional fundraising in order to continue their exploration into GenAI.

The Vision

This nonprofit organization had ambitious goals of leveraging AWS services to enhance their learning experience with Generative AI. The idea is that students can follow the course curriculum the same way they do currently, but receive additional feedback and recommendations for additional exercises and content based on their performance on assignments.

The Goal

This nonprofits courses teach about various technologies using the same curriculum throughout the entire course for every individual student. This is fairly standard in the learning industry, however they wanted to improve the student experience, by honing in on specific areas that they may need additional support with. By using GenAI to automatically provide recommendations based on student progress, the ideal outcome will be a higher success rate, as well as satisfaction from students in how well they learned the course content.

The Challenge

Instructors face difficulty identifying personalized extracurricular assignments due to the volume of students and varying performance across subjects. Manually curating content for each student is infeasible and time-consuming.

The Solution

As an instructor looking to assign after-class assignments to students with customized curriculum based on their historical class performance, this POC shows the feasibility of doing so using Amazon Bedrock. First, the student’s gradebook is provided in the request, and the tool will assign them a value based on their performance on various subjects:

  • Low: build foundation - re-teaching and summarizing
  • Medium: practice - reinforce and address partial information
  • High: extend - extend and apply knowledge

Then, it will find exact matching resources, and summarize according to what the student will need in order to reach a mastery level. Finally, the tool will provide its response with a cited resource in DynamoDB that is used to link to the knowledge base documents in S3. What this does, is determine what course content a student needs in order to improve on any particular subject based on their previous performance on assignments in the students gradebook.

OUTCOMES

Your data is trying to tell you something

Contact us

... are you listening?