Whatnot and Kumo.AI Leaders Discuss the Latest in Machine Learning [Videos]

Whatnot Engineering
Whatnot Engineering
3 min readMar 16, 2023

--

Machine learning powers countless areas at Whatnot — including onboarding, discovery, recommendations, and maintaining a relevant user journey.

Recently, leaders from Whatnot and Kumo.AI came together to discuss the latest in machine learning and AI, the power of graph neural networks, and the future of deep learning in enterprise.

Speakers included:

  • Whatnot Machine Learning Manager Ford Bohrmann
  • Kumo.AI VP of Product Tin-Yun Ho
  • Kumo.AI Co-founder and Chief Scientist Jure Leskovec

The night closed with a Q&A with the speakers hosted by Whatnot Community Engineering Lead Sophia Feng, along with Kumo.AI co-founder and Head of Engineering Hema Raghavan, and Whatnot Head of Machine Learning Emmanuel Fuentes.

For those who missed the event — or want to relive it — here’s an overview of the evening, complete with videos. Enjoy!

A Deep Dive into Machine Learning at Whatnot

Ford discussed feed ranking opportunities and challenges when building a livestream shopping marketplace. Prior to Whatnot, Ford led a team of Research Scientists at Meta.

The Future of Enterprise and Deep Learning

Tin-Yun shared the latest on the Kumo Graph ML Platform. Before joining Kumo.AI, Tin-Yun was a founding PM at Google Cloud, where he served as co-lead on Vertex, Google Cloud’s flagship ML platform.

The Power of Graph Neural Networks

Jure discussed graph neural networks and how it’s transforming the future of ML and predictive analytics. Prior to co-founding Kumo.AI, Jure was Chief Scientist at Pinterest and Co-founder of Kosei. He also serves as an associate professor of Computer Science at Stanford University.

Q&A: Fueling Growth with ML & AI

The group discussed growth loops (activation, finding relevance, retention) — including topics such as personalization in onboarding (balancing explore vs. exploit), maintaining fairness over biases, evolving user preferences, cohort detection, and data science and product collaboration.

We hope to see you at our next event! Stay connected to us on our blog and LinkedIn.

If you’re interested in working with us, check out our open roles.

--

--