Build & maintain complex distributed systems
October 1–2, 2017: Training
October 2–4, 2017: Tutorials & Conference
New York, NY

Putting your first paper into production

Ben Linsay (Bumpers)
1:30pm2:10pm Tuesday, October 3, 2017
Distributed Data & Databases
Location: Regent
Average rating: **...
(2.50, 2 ratings)

Who is this presentation for?

  • Engineers, data scientists, and anyone building data products or using machine learning

Prerequisite knowledge

  • A basic understanding of engineering and machine learning

What you'll learn

  • Learn how Kickstarter solved problems with machine learning

Description

Machine learning is as accessible as it has ever been, but it’s not always obvious how to go from a cool paper to serving production traffic. Ben Linsay helps you get started putting your paper into production, sharing lessons learned solving real problems with machine learning at Kickstarter.

Ben covers tips for identifying a solvable problem, picking a dataset responsibly, running offline experiments, and not getting stuck in the prototyping phase before outlining the infrastructure that worked for Kickstarter, which ranges from a Ruby gem to a two-tier distributed system. Along the way, Ben highlights the trade-offs made, emphasizes the importance of measurement at each step, and discusses why focusing on a specific problem to solve and being diligent about measurement led to success.

Photo of Ben Linsay

Ben Linsay

Bumpers

Ben Linsay is an engineer. In past lives, he was an engineer at Bumpers, Kickstarter, Aggregate Knowledge, and Boundary.