Back to feed
News Story
The AI Insider
1 sources

Whatnot Acquires AI Recommendation Startup Shaped to Boost Live Commerce Personalization

Whatnot, a livestream shopping platform, has acquired Shaped, a machine learning startup specializing in real-time recommendation and search technology. The acquisition aims to enhance personalization for live commerce, where inventory and buyer intent shift rapidly. Shaped's founder and nearly a dozen engineers will join Whatnot to form an Applied AI Research group.

SynthePulse Insight · AI deep reading

Whatnot Acquires Shaped: How Real-Time Recommendations Are Redefining Personalization in Live Commerce

Version 1 · 1 source

Live commerce platform Whatnot acquires AI recommendation startup Shaped, aiming to push recommendation latency from minutes to near real-time, addressing the constant shifts in inventory and buyer intent during live shopping. This deal is not just a technology integration but marks a critical shift in live commerce personalization from 'historical behavior analysis' to 'real-time intent response.'

  • Whatnot acquired Shaped, a machine learning company specializing in real-time recommendation and search technology, to enhance personalization in live commerce.
  • Shaped's technology combines customer data with large language models and previously served clients such as Outdoorsy and QVC.
  • Whatnot's VP of Data and AI stated that the company spent six years reducing recommendation latency from about one day to minutes, and aims to move closer to real-time by integrating Shaped.
  • Whatnot's system processes over 500,000 hours of live video and millions of interactions weekly to optimize recommendation models.
  • Shaped founder and CEO Tullie Murrell and nearly ten engineers will join Whatnot, with Murrell leading the newly formed Applied AI Research team.
  • Whatnot recently surpassed 1 billion orders and completed a $225 million Series F funding round at a valuation of over $11 billion.
Open section navigationAcquisition Background and Deal Details

Acquisition Background and Deal Details

On July 27, 2026, live shopping platform Whatnot announced the acquisition of machine learning company Shaped, which specializes in real-time recommendation and search technology. Shaped's technology previously served clients such as Outdoorsy and QVC, with its core capability being the combination of customer data with large language models to enable personalized search and discovery. As part of the deal, Shaped founder and CEO Tullie Murrell and nearly ten engineers and researchers will join Whatnot, with Murrell leading the newly formed Applied AI Research team.

Technology Integration Goal: From Minutes to Near Real-Time

Whatnot's VP of Data and AI, Emmanuel Fuentes, stated that integrating Shaped's technology will make recommendations faster and more responsive, addressing the unique challenges of live commerce—where inventory and buyer intent constantly change during auctions lasting minutes or hours. Fuentes revealed that Whatnot has spent six years reducing recommendation latency from about one day to minutes, and aims to move closer to real-time by integrating Shaped. Currently, Whatnot's system processes over 500,000 hours of live video and millions of interactions weekly to optimize recommendation models.

Whatnot's Growth Trajectory and Strategic Intent

This acquisition comes after a period of rapid growth for Whatnot. The company recently surpassed 1 billion orders and completed a $225 million Series F funding round at a valuation of over $11 billion. Acquiring Shaped aims to strengthen its personalization capabilities to support category expansion and a growing buyer base.

Credibility boundary

The information in this article is primarily sourced from a report by The AI Insider, which cited public statements from Whatnot executives and company-disclosed data. All figures and statements come from this single source and have not been independently verified by third parties.

Insight takeaway

The core logic behind Whatnot's acquisition of Shaped is that personalization in live commerce must shift from 'historical behavior analysis' to 'real-time intent response.' Shaped's real-time recommendation technology and large language model capabilities, combined with Whatnot's existing minute-level recommendation system and massive live-streaming data, are expected to push recommendation latency toward near real-time, enabling more precise matching of supply and demand in dynamic live scenarios. This deal also reflects the trend of AI recommendation technology evolving from general optimization to domain-specific models in vertical scenarios like live shopping.

Primary report

The AI Insider

Primary source