Customer Story:

Yotpo's ML models productionized with Qwak  

Yotpo is a pioneer in innovative ecommerce technology that leads to breakthrough moments throughout the creation process. On a mission to push the limits of technology to reimagine the way creators express themselves, the company brings a unique blend of cutting-edge academic research and design to every user experience.
Yotpo was considering building their own platform however after researching the options Yotpo decided to focus on their business and onboard Qwak which addresses Yotpo's requirements and has the required flexibility to support Yotpo for the long run.

Millions
End user
recommendations
0
Engineering
dependency
Full
Platform
Flexibility
About Yotpo
Yotpo is an eCommerce marketing platform that helps brands drive growth by creating engaging experiences to build lasting customer relationships. Their integrated solutions for reviews, visual User Generated Content (UGC), rewards, and referrals empower businesses to win over new audiences using their customers' voice.
Industry
Ecommerce marketing
Social Media
Adtech
Use Case:
Recommendations engine on user feed page
‍‍
Model Frameworks:
Pytorch
Qwak stack:
Build system
Serving/Hosting
Analytics
Feature Store
"Our data science teams deliver end to end ML model services however building infrastructure is not our business focus and therefore Qwak were ideal for us"
Jonathan Yaniv
Head of Data Science @ Yotpo
The Challenge
Developing recommendations engine models required intensive engineering support and building of new infrastructure layers that can scale and support billions of users in serving.
  • Needed a managed platform for Realtime ML model serving platform that is reliable, easy to use, can scale and is future proof.
  • The platform's flexibility was crucial. Yotpo did not want to be locked down to a specific model structure and also needed a centralized way to train, deploy and access models at scale.
  • Building Infrastructure requires a lot of effort, experience and time and is not Yotpo's business focus.
See it in Action
How Qwak Assisted?
  • Qwak Build The initial integration of Qwak required wrapping the Python model and took only a few simple lines of code to allow Yotpo to run a deployed system. Beyond the list of features that come Out of the box, Qwak's platform is flexible and designed to allow fast implementation of new functionality.
  • Qwak Console A fully functional UI allows Yotpo to have full visibility of all the deployed models and integrate the models to Opsgenie which is Yotpo’s central alerting system.
  • Qwak Serving provides simple, scalable deployment of real-time & batch models with the ability for gradual deployment (canary deployment) which allows Yotpo to support millions of predictions in real-time.
  • Qwak feature store  provides a simple API  for feature extraction retrieval of features for training. This allows the data science team to work autonomously and modify model inputs without the need to make backend modifications. It also provides the ability to share feature sets between different models and train multiple models at scale.
  • Qwak's Analytics allows reviewing the model predictions data and creating dashboards on top of the real time data to provide model insights for the business users.
"Qwak is built in a flexible way. Out of the box functionality is great but we were really glad to see that the platform was developed with flexibility and customizability in mind. That ensures that we can depend on it for the long term" 
Jonathan Yaniv
Head of Data Science @ Yotpo

Learn More About The Qwak Platform

 
Learn More