Dr. Kamelia Aryafar

Paul Anderson photo

Chief Customer and Algorithms Officer

As Overstock.com’s Chief Customer and Algorithms Officer, Dr. Kamelia Aryafar leads the company’s
machine learning, data science, data engineering and analytics functions across the marketing,
customer, sourcing, and website verticals. Since joining Overstock.com in 2017, Dr. Aryafar’s teams have
integrated state-of-the-art machine learning and artificial intelligence algorithms across various product
teams, including personalization, pricing, ranking, search, recommender systems, marketing, CRM,
advertising technologies, email, sourcing and supply chain.

With the ability to leverage two decades of customer data, Dr. Aryafar is driving a unique strategy at
Overstock. By applying her expertise in machine learning, she is amplifying even greater customer-
centricity across Overstock’s web and mobile experiences. From the algorithms that power the search
experience to the data used to personalize that experience, she is helping customers discover the most
relevant products and information to meet their needs and inspire their wants.

Aryafar transitioned from academia to the industry as a senior machine learning scientist and lead at
Etsy in 2013, where she worked with different product teams across several platforms to integrate
machine learning and artificial intelligence algorithms throughout the organization. This work allowed
her to apply the latest technology in the realm of e-commerce.

Dr. Aryafar holds a Ph.D. and M.Sc. in computer science and machine learning from Drexel University
and has published several papers in scientific journals. As an active member of the machine learning
community, she frequently speaks at academic and industry conferences focusing on advancing the field
of artificial intelligence.


Dr. Kamelia Aryafar is speaking at the following session/s

Keynote Panel: Building and Running Solr at Scale

Thursday | 9:30AM - 10:05AM | International Ballroom

Panelists will discuss what it takes to run Solr at scale for their business, including deployment tradeoffs (on-prem versus cloud), how to build a team to support Solr at scale, and developer success and how to stoke app development. They'll also share tricks of the trade including plug-ins, connectors, and integration points, as well as dreams for the future of Solr and how to measure success.

The audience will leave with an understanding of how to successfully set up a team to deploy, monitor and run a large-scale Solr stack and how to make it useful to the broadest possible audience. 

All Levels

Keynote: Where is my Sofa?! The Power of Multimodal Learning to Rank in Production Scale E-Commerce Search

Thursday | 10:05AM - 10:30AM | International Ballroom

Search is at the heart of modern day e-commerce. In this talk, Dr. Aryafar will share how deep learning and multimodal learning to rank methods can improve the relevancy of production scale e-commerce search engines. Kamelia will also share best practices on building successful AI teams and an experiment-driven culture, including how Overstock.com approaches search and machine learning and how she sees the field evolving.

AI in Practice: Separating Fact from Fiction

Wednesday | 5:00PM - 6:00PM | International Ballroom

Grant Ingersoll, CTO and founder Lucidworks, assembles some of the best minds in AI and puts them on the spot. This real-world look at AI goes beyond the hype to answer honest and thoughtful questions about data science, machine learning and NLP in practice. We’ll hear about both the successes and the failures. What can you learn from these less than stellar experiences?  When do you need AI – and when can other options suffice? And how do you assemble the right people to help transform your company? Can you retrain those you have?

Joining Grant to answer these and other questions are Kamelia Aryafar, Nick Caldwell, Dr. James Shanahan, and Alice Albrecht.


All Levels