Self-service data analytics powered by GPT-3
Acolyte is self-service business analytics. Users can ask questions like "how many customers do we have?" or "who placed the most orders last week?" in plain English and get the right answer from their data warehouse.
There are two separate applications to deploy (the api and client). The easiest way to deploy is on Heroku via the following steps:
- Create a new API Heroku app and a new Client heroku app.
- In the API app (via the Heroku command line or the Heroku web UI):
- add the
heroku/python
buildpack. - Add a
DATABASE_URL
environment variable with the full URI to your database (i.e.postgres://USER:PASSWORD@HOSTNAME:PORT/DATABASE
) - Add an
OPENAI_API_KEY
environment variable with your OpenAI API key.
- add the
- In the Client app (via the Heroku command line or the Heroku web UI):
- Add the
heroku/nodejs
buildpack and thehttps://github.com/heroku/heroku-buildpack-static
buildpack. - Add a
VUE_APP_API_ENDPOINT
with the url to your API (e.g.https://acolyte-api.herokuapp.com/api
) - Optionally, if you want to track usage using Heap, add a
VUE_APP_HEAP_ID
environment variable with your Heap environment ID.
- Add the
- In this directory, via the Heroku command line, run the following:
heroku git:remote --remote heroku-client -a YOUR_CLIENT_HEROKU_APP_NAME
heroku git:remote --remote heroku-server -a YOUR_API_HEROKU_APP_NAME
git subtree push --prefix client heroku-client master
git subtree push --prefix api heroku-server master