Presented By O’Reilly and Cloudera
Make Data Work
March 5–6, 2018: Training
March 6–8, 2018: Tutorials & Conference
San Jose, CA

Approaching the pricing problem at Lyft

11:50am12:30pm Thursday, March 8, 2018
Average rating: ***..
(3.00, 3 ratings)

Who is this presentation for?

  • Data scientists and business analysts

Prerequisite knowledge

  • A basic understanding of statistics

What you'll learn

  • Learn how Lyft approaches the pricing problem for a real-time, strategic perspective


Ashivni Shekhawat explains how Lyft uses a mix of online learning, optimization, and control theory to operate its ride-sharing marketplace at an efficient price point. Ashivni touches on various aspects of model development, experimentation, and testing in a marketplace with strong interference, inference in scenarios with data sparsity and class imbalance, and developing scalable infrastructure for training and deploying models.

Ashivni Shekhawat


Ashivni Shekhawat is a data scientist at Lyft working on pricing. Ashivni has developed several algorithms for dynamic pricing, online learning and estimation at Lyft. Ashivni is deeply interested in statistical inference, experimentation, and machine learning. Ashivni comes from a physics and engineering background, and as conducted research and graduate studies at UC Berkeley, Cornell, and IIT Kanpur. He holds degrees in Aerospace Engineering, Physics and Applied Mechanics