Skip to content

An example of using dbt to enrich data in BigQuery with LLM model

Notifications You must be signed in to change notification settings

pilis/dbt-bigquery-llm-enrichment-example

Repository files navigation

dbt-biguqery-llm-enrichment-example

This is an example of using dbt to enrich data in BigQuery with LLM model.

Getting started

Prerequisites

  • Python 3.11
  • GCP project

Setup

GCP setup

Create a connection of Cloud Resource type in BigQuery:

Replace YOUR_REGION and YOUR_PROJECT_ID with your own values

bq mk --connection --location=YOUR_REGION --project_id=YOUR_PROJECT_ID --connection_type=CLOUD_RESOURCE cloud_resources_connection

call_llm_model Cloud Function setup

Please follow the instructions in remote_functions/call_llm_model/README.md

dbt setup

Create a virtual environment:

python -m venv venv
source venv/bin/activate

Install dependencies:

pip install -r requirements.txt

Setup dbt profile in ~/.dbt/profiles.yml

Test the connection:

dbt debug

Run dbt models:

Before running the models, please make sure that you have created a connection of Cloud Resource type in BigQuery and that you have created a Cloud Function call_llm_model in your GCP project.

dbt build

Authors

Special thanks to Piotr Chaberski for reviewing the code and providing valuable feedback.

License

This project is licensed under the MIT License - see the LICENSE file for details

About

An example of using dbt to enrich data in BigQuery with LLM model

Topics

Resources

Stars

Watchers

Forks

Languages