

RecSys in production: Faire x Paramount x Anyscale
Recommendation systems (RecSys) are the revenue engine of consumer products. They decide what gets seen, bought and considered to grow revenue and increase customer stickiness.
But the stack behind them keeps getting more complicated. Classical ML rankers and two-tower deep learning models now run alongside LLMs, and personal agents now are also entering the loop.
Join ML engineers from Faire, Paramount and Anyscale for an evening of practical talks on how teams scale RecSys workloads including training, embeddings, and inference with mixed model workloads.
Then stay for food, drinks and time to talk shop.
Agenda
5:30 PM – Doors open, food, and networking
6:00 PM – Faire: Matt Wheeler, Infrastructure Engineer, ML Platform
6:30 PM – Skydance/Paramount: Adam Leary, Senior VP ML Engineering
7:00 PM – Open networking + happy hour
8:30 PM – Event Ends
Hosted by
Anyscale, the company behind the Ray open source project, is a platform that AI teams use to run data, training, inference, and RL workloads at foundation-model-scale on any cloud.
Faire is the online wholesale platform connecting retailers and brands. Retailers discover unique products from over 100,000 brands, while brands reach new customers and grow their business.
Skydance unites two major film studios - Paramount Pictures and Warner Bros., two major streaming services - Paramount+ and HBO Max, plus live sports and deep library of iconic franchises including CBS, HBO, CNN, and more.