Interested in this use case?

If you'd like us to focus on building this template next, register your interest and let us know. You can also head over to the Quix Community Slack if you've got any questions.

Built on Quix with:

Project template

Use case

Code snippet

Real-time feature computation

A real-time feature pipeline that computes essential features (open-high-low-close prices) that you can pass to an ML-powered trading system for training and inference. This template was contributed by Pau Labarta Bajo.

Use cases:

Feature Engineering

Created by:

Paulescu

A diagram of a real-time feature pipeline

Main project components

Trade producer

A source connector that reads Bitcoin trades from the Kraken Websocket API and writes them in a Kafka topic.

Trade to OHLC

A transformation that reads trading data from a Kafka topic, computes Open-High-Low-Close candles (OHLC) using Stateful Window Operators, and saves them in another Kafka topic.

OHLC to feature store

A sink connector that saves the computed OHLC data to Hopsworks—an external feature store.

Streamlit Dashboard: last 20-minutes

A dashboard that visualizes the last 20 minutes of the real-time OHLC data based on trading activity in Kraken (computed on 10-second windows)

Technologies used

Using this template

This project could be easily adapted for use cases such as:

More advanced trading analysis
Other window-based pipelines such as temperature analysis

Interested in this use case?

If you'd like us to focus on building this template next, register your interest and let us know. You can also head over to the Quix Community Slack if you've got any questions.

Built on Quix with: