hayaData Meetup - How to Develop and Deploy Dozens of ML Models at Scale

Join us for the first hayaData meetup, presented by Riskified Technology and Fiverr, about handling, monitoring and retraining multiple ML models at scale.

Agenda:
18:00 - 18:30 - Mingling, Beer and Snacks.
18:30 - 19:00 - Managing Multiple ML Models For Multiple Clients: Steps For Scaling Up.
19:00 - 19:30 - How We Built Our ML Platform from Scratch.
19:30 - 20:00 - More Beer and Mingling.

The talks will be in Hebrew.
This is an in-person event and will take place in accordance with the Green Pass.
*Due to the uncertain situation caused by COVID-19, the event might change to another date.
We will provide relevant updates here.

------------
//Managing Multiple ML Models For Multiple Clients: Steps For Scaling Up - Ori Peri, ML Engineer, Riskified.

For most ML-based SaaS companies, the need to fulfill each customer’s KPI will usually be addressed by matching a dedicated model. Along with the benefits of optimizing the model’s performance, a model per customer solution carries a heavy production complexity with it. In this manner, incorporating up-to-date data as well as new features and capabilities as part of a model’s retraining process can become a major production bottleneck. In this talk, we will see how Riskified scaled up modeling operations based on MLOps ideas, and you will receive tools for how to set up your own continuous training ML pipeline.

Ori is an ML Engineer, part of the team responsible for building and operating Riskified’s MLOps tools, including the analytical and technical sides of maintaining production models and their (re)training process validation. He has 5 years of experience in multiple SW and ML engineering roles. He is passionate about SW, ML and incorporating one into the other.

Linkedin: www.linkedin.com/in/ori-peri

//How We Built Our ML Platform from Scratch - Lior Tzur, DevOps Engineer and Or Levi, ML Engineer, Fiverr.

Machine Learning has become an integral part of Fiverr’s marketplace; from giving personalized experiences to buyers to detecting spam messages sent in their system.
In this talk, you will see how Fiverr’s generic platform for training and deploying models works and allows scaling the development of dozens of ML models.

Lior has 10 years of experience in the DevOps, ALM and infrastructure worlds. He started his career at the IDF, where he served for 8 years as an officer in several elite technological units, such as Mamram. At Fiverr, Lior is the DevOps lead of the ML platform, he built this ecosystem from scratch, which serves their data scientists on a daily basis and allows the development and deployment of all of Fiverr’s ML models.
Lior’s aspiration is to automate everything and help developers focus on development by providing robust DevOps solutions.

Linkedin: www.linkedin.com/in/lior-tzur

Or has 10 years of experience as a Software Engineer and with Big Data technologies.
He started his journey at Fiverr 4 years ago and was promoted to lead the development of Fiverr’s ML Platform. He served in the IDF Intelligence force (9900) and worked with Big Data technologies, processing millions of records.
Or is a Big Data enthusiast and loves distributed systems.

Linkedin: www.linkedin.com/in/orlevii