Customer Story:

How Qwak enhanced Spot by NetApp’s Data Science delivery

Spot by NetApp delivers unique technology for cloud automation and optimization, and empowers companies to get the most out of their cloud investments through world-class CloudOps.
Qwak was brought aboard to help enhance the team’s ML model to production time and eliminate the friction between data science teams and machine learning engineering.
Spot invests in Data Science and Machine Learning teams to dramatically enhance user value and generate predictions for infrastructure and cloud cost changes.
5X
Faster time to ML model deployment
85%
Less issues in production
5
New models running in production within 4 weeks
About Spot by NetApp
Spot by NetApp is changing the way that companies run in the cloud. Spot's product suite uses innovative machine learning and analytics to automate and optimize cloud infrastructure, ensure that workloads and applications always have the best possible infrastructure--always available, always scalable and always at the lowest possible cost.
Industry
Saas
Cloud Operations
Use Case:
XGBoost based on Elasticsearch data
Linear regression based on S3 data with daily retrain
Qwak stack:
Build system
Serving
Analytics
Feature store
Automation
“Qwak helped us to make a paradigm shift to our data science operations and put us in a place we were trying to get to for a while now. I feel that the real value in Qwak is the fact that it actually ensures that your models make it into production really fast. We now deliver new models quickly and efficiently and with much less friction along the process”
Amiram Shachar
Vice President & General Manager, Spot by NetApp
The Challenge
Deliver Models to Production efficiently and in a timely manner
  • Enhance Data Science version deployment with less engineering and/or DevOps dependencies
  • Test model versions and feature creation without impacting production environments
  • Centrally monitor model performance metrics and control changes. Align feature data pipeline for training and serving without causing data skews and team communication breakdowns
How Qwak Assisted?
  • Qwak Build standardizes an ML delivery project structure This allows every team member to receive the same code structure, data versioning and testing capabilities for every model version.
    Using
    Qwak Build increases team members' confidence that new versions won't break any production service and ensures the ability to rollback to previous versions at any time.
  • Qwak Hosting, allows Spot’s team to deploy a new real-time model with a click of a button and automatically support Spot’s deployment strategy.
  • Qwak Serving, allows Spot’s team to deploy a new real-time model with a click of a button and automatically support Spot’s deployment strategy which is to deploy a canary version that only a subset of users gets as the first part.
  • Qwak Analytics stores all the inference (prediction) data automatically and provides the  ability to query real production data at any stage
  • Using the Qwak Feature Store gives Spot’s Data Science teams the ability to manage their  own data pipeline while writing and scheduling their own features. This eliminates the dependency on engineering teams.By using the features that we built within the feature store we’re now able to train our models and serve them with ease. Moreover, we now have visibility for all of our feature freshness, lineage and one simple catalog that allows us to easily connect features to models.
“Before Qwak, delivering a new ML model took weeks, I needed to get the attention of the engineering and product teams for every minor change. Now the research team can work independently and deliver while keeping the engineering and product teams happy by providing high-quality standards and high visibility of the model behavior.”
Idan Schwartz
Head of Research, Ph.D. Deep Learning, Spot by NetApp

Learn More About The Qwak Platform

 
Learn More