Back to blog
End of Year Feature Announcements 2022

End of Year Feature Announcements 2022

Pavel Klushin

We are extremely excited to announce Qwak’s new ML model management and feature store enhancements. These new capabilities allow Qwak’s users to optimize ML model performance and deliver top grade machine learning production workflows.   

Machine Learning Model Management Enhancements

Experiment tracking - Side by Side Build Comparison

Qwak’s enhanced capability lets you organize, track, compare & evaluate ML experiments (‘builds’ in Qwak’s terminology) and model versions.

Experimenting is a crucial piece to deliver quality production ML models; however, comparing experiment results can become a tedious task and is often non conclusive due to lack of tracking capabilities.  This makes decision making very difficult.

Qwak users can now visualize and compare model parameters and metrics side-by-side for multiple ML model builds. Our new visualization of experiment tracking results empowers ML practitioners to explore numerous build combinations and identify the best performing models at a glance

Side by Side Model Comparison Chart

Machine Learning Feature Store Enhancements

UI driven ML feature creation

Qwak’s Feature Store now provides an intuitive visual feature creation interface, making it easy to build and ingest features. 

UI Driven Feature Creation on Qwak Feature Store

Qwak allows creation of new feature sets via an intuitive interface that reduces unnecessary complexities.

Feature set definition stage

Define the data source, entity, type, names and more. Select your feature set type whether categorical or aggregation and you're done.

Feature Quality Monitoring

Qwak’s new Feature sets descriptive statistics view provide an efficient means to summarize and quantify the properties of a feature set and provide information such as its central tendency (mean, median, mode) dispersion (range, variance, standard deviation). 

Descriptive stats enable quick and accurate information extraction from any given dataset. Knowing how to effectively interpret descriptive statistics can provide invaluable insights and help inform key decisions. 

To see the above functionality in action and more, please check out our free trial

About Qwak:

Qwak’s ML Platform and Feature Store empower data science and ML engineering teams to continuously build, train and deploy ML models to production at scale.

By abstracting the complexities of model deployment Qwak brings agility and high-velocity to all ML initiatives designed to transform business, innovate, and create a competitive advantage. Qwak is a fully managed MLOps platform that allows data teams to focus on the science and create a business impact.

To learn more about Qwak please visit or schedule a demo with us

Related articles