/ai-marketing-emails

Primary LanguagePython

AI Marketing

  • app : Streamlit app that integrates with Snowflake to generate prompts for OpenAI ChatGPT API and write the results back to Snowflake tables.
  • app/main.py: App Homepage
  • app/extraordinary_events.json: JSON file with system prompts for email generation.
  • app/pages/1_Generate_E-mails.py: Streamlit page with interface for crafting prompts and reading generated emails
  • app/pages/2_Edit_Data.py: Streamlit page for editting persistent app data.
  • app/pages/4_Browse_Log.py: Streamlit page that renders history of all prompts generated.
  • sql/setup.py: Python script that uploads extraordinary_events.csv to Snowflake account and creates useful UDFs.
  • aimarketing/: handful of utility functions.
  • environment.yml: environment for streamlit app. Limits to Snowflake repository so app could be run in Snowflake.
  • pyproject.toml: Allows pip install -e . for utility functions.

Setup Environemnt

conda env create --file environment.yml --prefix ./.venv
conda activate ./.venv
pip install -e .

Setup Snowflake Connector

Create ~/.snowsql/config

[connections.aws]
accountname = ACCOUNT_IDENTIFIER
username = USERNAME
dbname = SANDBOX
schemaname = AI_MARKETING
warehousename = PHDATA
rolename = DATASCIENCE
private_key_path = /path/to//rsa_key.p8

If you want to authenticate with password, replace private_key_path with password. The Snowflake Python connector expects an environment variable "PRIVATE_KEY_PASSPHRASE" to use the specified private key.

Run

export OPENAI_API_KEY="sk-****"
python sql/setup.py
python -m streamlit run app/main.py