Nixon Dinh is a product leader whose work connects agentic commerce with the practical needs of merchants: helping AI assistants find suitable products and turning those recommendations into purchases. In 2026, he served as PayPal’s director of product for agentic commerce, focusing on a shift from shoppers searching for items themselves to software interpreting their needs and acting on their behalf.
Making products understandable to agents
Dinh frames that shift as a progression from keyword search to conversational intent and then delegation. A shopper can describe a need rather than supply the words a retailer happens to use in its catalog. An agent must connect that request to a suitable product, making the quality of merchant product information central to the shopping experience.
His approach to catalog enrichment emphasizes catalog quality over text volume. PayPal’s experiments found that enriching merchant product data improved agent recommendations, with thin catalogs benefiting most. Adding excessive unstructured boilerplate, however, diluted useful information and could lead agents to hallucinate. The useful work is supplying details that help an assistant choose: more words can make that decision harder when they obscure the product’s relevant attributes.
Dinh also favors hybrid search, combining explicit keyword matching with semantic matching. The combination accommodates both precise product language and requests expressed in terms of a shopper’s needs, without expecting either retrieval method to handle every query.
Connecting discovery to a completed purchase
The merchant side of this work extends beyond recommendation quality. PayPal’s Store Sync connects catalog discovery, cart creation, and purchase completion with delivery of orders into a merchant’s existing order-management system. That connection matters because an agent’s product selection must eventually become an order the retailer can process and fulfill.
Dinh’s writing about the Feedonomics–PayPal Pacsun integration places a concrete retailer within this product picture. The integration connects merchant catalog information to agent-led shopping and the transaction path that follows. It is collaborative work, bringing together the catalog and commerce systems needed to make a new shopping channel usable.
Trust and the people building the products
Two further concerns broaden Dinh’s product perspective:
Transaction identity and liability: His position on trust in agentic transactions centers on the responsibilities introduced when software acts for a shopper. Identifying the participants and determining responsibility for a transaction become part of the commerce problem alongside finding and buying the right item.
AI-assisted prototyping by domain experts: Dinh advocates using AI to help domain experts build prototypes. People who understand a business problem can use AI assistance to turn that knowledge into something tangible and testable. This extends his interest in AI from the shopping experience to how products are developed.
Nixon Dinh explains how shopping moves from keywords to intent and delegated action, then follows PayPal’s catalog-enrichment experiments to a practical finding: useful product detail improves recommendations, while excess text can make them worse.
Intent-driven shopping makes product context more important: agents must connect a shopper’s situation to suitable items, rather than depend only on matching keywords.
Hybrid search combines predictable literal matching with broader meaning-based retrieval; the recording does not disclose PayPal’s ranking implementation.
PayPal’s experiments found the largest enrichment gains in thin catalogs. Product attributes, deeper descriptions, buyer context and reviews can supply information those records lack.
Over-enrichment sometimes increased hallucinations and reduced performance. Assess the starting catalog and add relevant product information rather than boilerplate.
▶ Watch ↗