Customer Story:

Salt's API security gets a boost with Qwak's ML Platform

Salt Security provides an API protection platform designed to prevent attacks by leveraging machine learning and AI.
Qwak was brought aboard to help Salt's growing data science team become self sufficient and reduce dependency on Devops or engineering teams. The requirement was to scale model delivery to production without increasing engineering efforts.
Salt identifies API security vulnerabilities across both published and private applications and services. The team publishes findings, following responsible disclosure or preserving anonymity, so that the industry can learn and improve API security.

Real-Time
Model analytics &
monitoring
Zero
Dependency
on engineering
Built-in
Streaming
inference
About Salt Security
Salt delivers protection for APIs across build, deploy, and runtime phases. Salt Security combines complete coverage and an ML/AI-driven big data engine to provide context across all your APIs, stop attackers during the early stages of an attempted attack, and share insights to improve API security posture.
Industry
Cyber Security
Software/Technology
Use Case:
LSTM network model
Qwak stack:
Build system
Serving/Hosting
Analytics & Monitoring
“As our data science team and customer base grew, we found it challenging to move our new and more sophisticated models into production, even though we knew the updated ones were better. We couldn’t stay in that mode"
Elad Weiss
Data Science Team Lead @ Salt Security
The Challenge
Eliminate the need for ongoing engineering efforts for model productization
  • Rapid company growth allowed Salt to expand the Data Science team and create more and better ML models however deploying these models to production became a significant challenge.
  • AWS SageMaker implementation required an Infra team to be actively involved in every new model. Even experimenting during the PoC was challenging without an experienced engineer at hand
  • Data Science teams do not have engineering knowledge or experience
  • Integrating to Kafka as event based architecture was a major challenge as Salt needed endpoints to receive and output predictions as streams
See it in Action
How Qwak Assisted?
  • Qwak Build allows Salt's team members to use a single code structure for all models which allows Salt to deploy models in production at scale.
  • Qwak's terminology helps Salt make sense of all moving parts of the ML infrastructure. The interface makes it very easy even for non experts in the field. All functionality is also available through the API.
  • Qwak Analytics and Feedback loops let Salt's teams validate model behavior almost instantly after deploying to production.
  • Qwak Serving allows deploying model endpoints with a native integration to Kafka with a single click. Salt's data scientists and engineers can deploy their new model versions as endpoints that can receive and output predictions as streams.
“We had the data, and we solved the problem, Qwak allowed our data science teams to deliver the models into production with ease and efficiency”
Elad Weiss
Data Science Team lead @ Salt Security

Learn More About The Qwak Platform

 
Request a Demo