Analyzing Coinbase Order Book in Real Time

This project demonstrates a data pipeline using the Coinbase websocket feed to display and analyze their order book in real time.

Reference Commands

Build docker images

docker-compose build

Start the services

docker-compose up -d

Stop the data generator

docker stop data-generator 

Access the dbt cli

docker exec -it dbt /bin/bash

Access the Materialize cli

docker exec -it mzcli psql -U materialize -h materialized -p 6875 materialize

Ports

• http://localhost:8080/ Redpanda Console
• http://localhost:8000/ dbt Docs
• http://localhost:8501/ Streamlit App

Resources & References

These are resources and references I used when building this project.

• Real-Time Financial Exchange Order Book
• Streaming Data Apps with Bytewax and Streamlit
• How to Build a Real-Time Feature Pipeline In Python
• Temporal analysis of Wikipedia changes with Redpanda & Materialize & dbt
• Breathing life into Streamlit with Materialize & Redpanda
• How to build a real-time crypto tracker with Redpanda and QuestDB
• Online Machine Learning in Practice: Interactive dashboards to detect data anomalies in real time
• Materialize Top K by group
• Materialize + Redpanda + dbt Hack Day
• Pandas Table Viz
• Plotly Filled Area Plots
• Deploy Streamlit using Docker