qlikstar/sf-ai-hack-backend

MagicReply

Free yourself from replying repetitive emails. MagicReply learns from your past emails and your documents to help you automate email responses.

Inspiration

We are tired of replying emails. A lot of my knowledge exists somewhere, I shouldn't have to write emails from scratch every time.

What it does

It ingests knowledge from my past emails and documents, and synthesize a email response to answer incoming emails.

How we built it

RAG + Chrome Extension built on Javascript

Challenges we ran into

RAG latency, injecting responses to front-end

Accomplishments

Built something end to end

What we learned

Building a gen-AI App

This repo is the back end code for our bot-extension project, which you can find here

Running the app locally

1. Install virtualenv and activate it

   apt install python3-virtualenv -y

   virtualenv venv

   source venv/bin/activate

2. Install the requirements

   pip install -r requirements.txt

3. Create a file config.yaml with the OpenAPI keys by copying config.yaml.sample.

4. Run the service

   uvicorn app:app --host 0.0.0.0 --port 80