The problem LearnMateAI solves

-> LearnMateAI aims to solve the problem of traditional learning methods by adapting to the student's learning pace and style, focusing on important topics, and providing instant feedback through regular testing.

-> It is a personalized AI tutor that can help students learn a subject area in a more efficient and effective way.

-> The system will learn the subject area from the teacher’s notes and understand the important topics based on the previous year's question papers.

-> It will then teach the student step by step, focusing on the most important topics and adapting to the student's learning curve. Generate Imp topics and possible Question papers.

-> The system will interact with the student like an AI tutor who has learned all teacher notes and PYWS to create an engaging learning experience.

-> This personalized and adaptive learning experience can help students understand concepts better and achieve better grades.

Challenges we ran into

While building this project, we encountered a specific bug related to deployment. Initially, we attempted to deploy the system using Docker in the AWS environment, but we faced limitations with AWS Lambda regarding package imports, so we use the AWS layer but it also had the problem of 250 MB max packages as we are using TensorFlow and heavy ML packages it was not possible, then we tried AWS container integration with AWS lambda that was also not possible due to exceeding limit of the policy., it was almost 6 PM so rather on focus on live deployment we decided to make the actual thing first As a result, we had to change our approach and deploy the system locally instead.

Another hurdle we faced was the processing time of TensorFlow, which was taking longer than expected. To overcome this, we leveraged the Intel One API optimization tool(DNN). By utilizing the optimization capabilities provided by one API, we were able to enhance the performance of TensorFlow and significantly reduce the processing time.

These challenges required us to adapt our deployment strategy and optimize the performance of the system. Through careful troubleshooting and finding alternative solutions, we successfully resolved these issues and continued building the project