AssessMentor

Inspiration

We were inspired to create AssessMentor by our own college experiences studying for exams. We wanted to create a project that would improve students' everyday lives.

What it does

AssessMentor uses large language models to generate practice assessments to help with studying.

How we built it

AssessMentor creates a prompt to give to the LLM using AWS Bedrock. The response from the LLM is then used to create the practice assessment.

Challenges we ran into

We were new to some of the technologies that were used to make the project including AWS Bedrock.

Accomplishments that we're proud of

One of the biggest accomplishments was creating a polished Front-end with no heavy framework. We were able to implement smooth page transitions, responsive animations, and robust error handling.

We are also proud of being able to integrate multiple AWS services in a coherent way, from Bedrock for Generative AI to RDS for data persistence.

What we learned

We learned how to manage AWS credentials, integrate AI model APIs, design database schema, and optimize front-end for reliability and usability.

What's next for AssessMentor

We hope to implement Exam profiling, so that users can use customized presets instead of always manually inputting preferences. We also plan to build AI-based answer evaluation, specifically on the open-ended.

Built With

• amazon-web-services
• aurora
• bedrock
• css
• flask
• html
• javascript
• llm
• prompt
• python