Whatnot Acquires Shaped to Power Real-Time AI Shopping

On Wednesday, live commerce giant Whatnot announced its acquisition of machine learning startup Shaped to integrate real-time AI recommendations and search tools into its rapidly expanding livestream shopping platform.

This strategic move represents a major step forward in Whatnot’s ongoing AI investments. By acquiring Shaped, the company aims to tackle one of the most complex challenges in live commerce: delivering highly accurate product recommendations to users in real time, even as inventory, live auctions, and buyer behavior shift by the second.

Solving the Live Commerce Discovery Challenge

Unlike traditional e-commerce websites where product catalogs remain static for long periods, Whatnot operates a highly dynamic marketplace. Live auctions on the platform can flash by in seconds or stretch on for hours, creating a volatile environment that traditional recommendation algorithms struggle to navigate.

“By combining Shaped’s technology with Whatnot’s existing systems, we can make recommendations faster, more responsive, and more personalized,” Emmanuel Fuentes, VP of Data and AI at Whatnot, told TechCrunch. “That speed matters because live commerce is a uniquely hard recommendation problem. Inventory changes by the second, shows start and end continuously, and buyer intent shifts throughout a show.”

Shifting Recommendations From Days to Seconds

Whatnot has spent the last six years optimizing its recommendation engine, successfully reducing latency from nearly a full day to just a few minutes. Integrating Shaped’s specialized technology is expected to push these systems even closer to true, instantaneous real-time processing.

Currently, Whatnot’s infrastructure processes over 500,000 hours of live video and millions of real-time user interactions every week. The platform leverages this massive influx of data to continuously train and refine its recommendation algorithms.

Who is Shaped AI?

Shaped was founded to help enterprises build powerful, AI-driven recommendation systems. The company’s technology merges existing customer data with large language models (LLMs) and advanced machine learning to deliver deeply personalized search and discovery experiences. Prior to the acquisition, Shaped provided its services to prominent clients, including QVC and Outdoorsy.

As part of the transaction, Shaped founder and CEO Tullie Murrell—a Meta alumnus—will join Whatnot alongside nearly a dozen machine learning engineers and AI researchers. Murrell is set to lead Whatnot’s newly established Applied AI Research group.

Rapid Growth and the AI Resale Race

The acquisition comes during a period of massive scaling for Whatnot. Founded in 2019, the livestream platform recently revealed that its sellers have surpassed 1 billion total orders. Whatnot also secured $225 million in Series F funding, pushing its valuation past $11 billion after adding 20 million buyers over the past year.

To support this growth, Whatnot has aggressively expanded its inventory. The platform launched more than 35 new product categories last year—including art, golf, and vinyl—and introduced over 45 additional categories during the first half of 2026, with new subcategories debuting on a monthly basis.

Whatnot’s latest acquisition highlights a broader industry trend, as major resale and e-commerce platforms like eBay and Poshmark race to integrate advanced AI technologies to personalize the user experience.