Tau

Inspiration

Math is hard, but maybe it shouldn't be?

What it does

A completely serverless cloud managed agentic AI platform that provides the user with a plethora of tools to accelerate the learning process. Leveraged AWS Bedrock to create many powerful agents that are specialized in their respective math fields. Deployed data pipelines that leverage machine learning image recognition to help get your written papers into LaTeX on the cloud.

How we built it

We utilized React and the native AWS SDK to facilitate easy communication between user and our agents. We also provide easy to use wrappers for many of AWS managed services.

Challenges we ran into

Integrating Lambdas with Bedrock to create powerful agents that can utilize real world data.

Accomplishments that we're proud of

Deploying a machine learning model on SageMaker to be able to take a picture of papers on a desktop all the way to a pdf rendered with LaTeX. Also, integrating multiple cloud-managed services with bedrock agents.

What we learned

We learned how we can use Sagemaker AI to deploy models and use them with data pipelines. Also, how to create a fully serverless agent backend inside the AWS ecosystem.

What's next for Tau

We want to turn this into a more fleshed-out platform that allows users to create and easily manage their projects and lessons. Also, creating more tools to help those who are interested in learning.

Built With

• amazon-web-services
• bedrock
• go
• lambda
• react
• sagemaker
• terraform