MailMatic is a simple web app that helps you write perfect emails with ease. Submit a few notes and the AI will write an email for you.
Users without an account can submit notes of up to 150 characters, while registered users can submit notes of up to 300 characters. Additionally, users are limited to a maximum of five submissions per hour.
A demo can be viewed below. To access the application, click on this link: mailmatic.vercel.app.
Under the hood MailMatic is running OpenAI's GTP-3 model Davinci. The interactions with Davinci are handled by a serverless AWS backend written in Python. The frontend is built with NextJS and TailwindCSS written in Typescript and is hosted on Vercel. User authentication is handled via Firebase. Rate limiting is implemented using Redis.
In order to run the project locally, you will need:
- OpenAI API key
- AWS account
- AWS CDK set up
- Docker running
- Firebase account
- Redis DB
Put the respective credentials in a .env file analogous to backend/.example.env and mailmatic-site/.example.env.
Now you need to set up the backend infrastructure in AWS. First, generate base layer for AWS Lambda.
cd backend/infra/lambda_base_layer
chmod -x generate_base_layer.sh
./generate_base_layer
This should create a layer.zip file. Then deploy the stack.
cdk deploy
Once the AWS infrastructure has been set up, you can run the application and it will be available at http://localhost:3000
cd mailmatic-site
npm run dev
A big thank you to @pixegami for the helpful tutorials. Your videos helped me tremendously in this project.