Build Systems that Drive Business
Sep 30–Oct 1, 2018: Training
Oct 1–3, 2018: Tutorials & Conference
New York, NY

Migrating a recommendations platform from bare metal to the cloud

Leemay Nassery (Comcast)
1:30pm–2:10pm Wednesday, October 3, 2018
Distributed Data
 Location: Nassau Level: Intermediate
Secondary topics:  Systems Architecture & Infrastructure
Average rating: **...
(2.00, 1 rating)

Prerequisite knowledge

  • A basic understanding of managing data pipelines at a larger scale

What you'll learn

  • Understand the importance of data collection pipelines
  • Learn how Comcast migrated its recommendations platform from a bare-metal Hadoop infrastructure to an event-streaming cloud platform

Description

You cannot have a machine learning platform that drives content recommendations without data. You also cannot properly measure the performance of recommendations without data.

Data is the key to monitoring, improving, and building complex recommendations systems that impact UX experiences. Leemay Nassery discusses the importance of data collection pipelines and explains how to efficiently store datasets by describing how Comcast migrated its recommendations platform from a bare-metal Hadoop infrastructure to an event-streaming cloud platform. Leemay begins by exploring the legacy data platform and architecture in detail and comparing it to the current cloud-based system. She then explains how these changes increased reliability and stability and generally improved the results generated by the respective downstream consumers, like Comcast’s machine learning tier.

Photo of Leemay Nassery

Leemay Nassery

Comcast

Leemay Nassery is a senior engineer leading the recommendations and targeting engineering efforts at Comcast. She also sets the strategic direction for content personalization for Comcast’s Xfinity consumer-facing video products and leads efforts with A/B testing, testing and targeting, and producing the metrics to measure successful customer outcomes.