https://github.com/hegelai/sdr-demo

https://github.com/hegelai/sdr-demo

Last synced: 7 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/hegelai/sdr-demo
• Owner: hegelai
• License: apache-2.0
• Created: 2023-12-29T02:11:02.000Z (almost 2 years ago)
• Default Branch: main
• Last Pushed: 2024-01-09T19:12:39.000Z (almost 2 years ago)
• Last Synced: 2025-01-17T11:13:11.486Z (9 months ago)
• Language: TypeScript
• Size: 507 KB
• Stars: 3
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# SDR-Demo

> DISCLAIMER: This app is a minimal demo for [Hegel AI's](https://hegel-ai.com) monitoring and online evaluation platform. You can sign up [here](https://app.hegel-ai.com).

:wave:

Hey there, 

SDR-Demo is a GPT-powered email generator. It uses key-value pairs describing a prospect, and a text input description of the offer, to generate an outbound sales email.

![image](images/preview.png)

At the bottom of the generated email, users can provide a thumbs up or thumbs down response for the generated email, which will be recorded by Hegel AI's logger.

![image](images/feedback.png)

Logs will be visible at https://app.hegel-ai.com/logs.

![image](images/logs.png)

## Getting started

You can kick off the frontend by running `npm start` from the `frontend` directory. The app should start running on `http://localhost:3000`.

### TypeScript Node.js Backend

To kick off the backend, you must set your `HEGELAI_API_KEY` and `OPENAI_API_KEY`. If you need a Hegel AI API key, you can sign up [here](https://app.hegel-ai.com). Then, you can run `npm i; npx ts-node app.ts` from the `node/` subdirectory to start the backend on `http://127.0.0.1:5000`.

### Python Flask Backend

To kick off the backend, you must set your `HEGELAI_API_KEY` and `OPENAI_API_KEY`. If you need a Hegel AI API key, you can sign up [here](https://app.hegel-ai.com). Then, you can run `export FLASK_APP=app.py; python -m flask run -p 5000` from the `flask/` subdirectory to start the backend on `http://127.0.0.1:5000`.

After starting the frontend and backend, the app should be working end to end. As long as your Hegel AI API key is set, logs will be visible at https://app.hegel-ai.com/logs.