Leading AI Innovation at Shopify: How We Built the Foundation for Shop App and Shop AI
TL;DR
Between July 2021 and May 2023, I led the Product Understanding team at Shopify, where we built the AI foundation that powers some of the company’s most innovative products. Our work enabled the Shop App’s intelligent navigation and Shop AI’s conversational shopping experience—demonstrating how strategic AI investments can transform entire product categories.
You can experience our work firsthand in Shop.App through category navigation, product search, and personalization features. Or chat with Shop AI to see how AI can understand and discuss products with human-like intelligence.
The Challenge: Building AI That Understands Products Like Humans Do
The Problem
Have you ever wondered what happens when you type $\sqrt{783225}$ in a handheld calculator? I mean, what marvelous algorithm runs this complicated task with so few resources and gives you the exact result in milliseconds? The answer is binary search, but that’s not the point. The point is that the best engineering projects are like this—they just work, and you never notice them until they break.
AI generated image for “binary search on abacus”
Online product discovery is similar. You never notice a good search engine until it starts returning bad results, and you never question category navigation until yellow shoes appear in the “white sneakers” category.
While Amazon requires GTIN IDs of the products 1 and Google Shopping requires using Google Taxonomy2, Shopify’s philosophy aims to reduce the complexity of setting up an online store, so anything not crucial to selling the product is optional. In addition, Shopify merchants are not bound by any taxonomy or guidelines, so they are free to use any taxonomy and product attributes that work for them.
With the lack of restrictions, it’s easy to imagine how a merchant can tag their yellow shoes with white embellishment with both “yellow” and “white” tags to improve SEO. This is fine for the online store, but when creating a cross-merchant experience like Shop.App, yellow shoes can easily end up in the white sneakers section or appear in search results for a “white sneakers” query.
Our Strategic Approach
To tackle this problem, we needed more than just better algorithms—we needed a fundamental rethinking of how AI understands products. I led our team in building a layer of ML-powered software that processes all unstructured and semi-structured data on products and maintains an up-to-date snapshot of calculated product metadata in a structured form with standardized language across the organization.
Building the AI Foundation
The Data Assets We Created
Our Product Understanding team focused on three critical types of metadata that would become the foundation for intelligent e-commerce:
Product Categories: What the Product Is
We built a hierarchical category tree, similar to Google’s taxonomy, that describes what the product is. An example of a product category from Google’s taxonomy would be Business & Industrial > Food Service > Hot Dog Rollers.
Product Attributes: What the Product Has
Product attributes describe everything the product has. These are key-value tags, such as Sleeve Length: Short. The attributes can apply to multiple types of products; in this example, it applies equally to tops, dresses, and jumpsuits.
A typical top has around 15 attributes calculated
Product Embeddings: The Nuanced Understanding
Product embeddings represent all other properties of the product that aren’t captured in categories and attributes. This includes vague properties like “summer dress” or “boyfriend jeans,” as well as properties that can’t be defined at all, such as “has a similar vibe to this product.”
The Technical Innovation
While I cannot disclose how these models were trained, I can mention that, unlike the approach taken at Donde Search, which focused mostly on automating supervised learning techniques and speeding up the manual annotation process, at Shopify we leveraged different NLP methods, zero-shot and few-shot models, to minimize the dependency on manual annotation.
Our team created dozens of models that have been deployed to Shopify’s production environment. These models run in a data pipeline on billions of products daily, operating efficiently by only predicting on individual images/texts that require an update. The pipeline handles all possible edge cases, such as a single image being changed on a product, a single text being changed on a product, or a new model being released and applying to a subset of the products.
The Impact: Transforming How People Shop
Shop App: Intelligent Discovery
The Product Understanding data assets power numerous teams and contribute to many Shopify products. However, I chose to focus on two products that use these assets in their rawest form.
You can experiment with category navigation on Shop.App or search for a specific type or attribute. But be aware that the relevancy of the search results is not accidental; our team spent countless hours perfecting the AI models that make this possible.
Shop AI: Conversational Commerce
Alternatively, you can try chatting with the Shop AI assistant to learn about available products. You will quickly notice just how deep its knowledge of the products is—this isn’t just keyword matching; it’s genuine understanding of product relationships, attributes, and context.
Leadership Lessons: Building AI Teams That Deliver
The Technical Challenge
Building AI systems that work at Shopify’s scale requires more than just good models—it requires:
- Data Pipeline Engineering: Systems that can process billions of products efficiently
- Model Management: Continuous deployment and monitoring of dozens of AI models
- Cross-Team Collaboration: Working with product teams to understand their needs
- Performance Optimization: Ensuring AI insights are delivered in real-time
The Strategic Vision
Our work wasn’t just about solving today’s problems—it was about building the foundation for tomorrow’s innovations. By creating standardized, AI-powered product understanding, we enabled:
- Faster Product Development: Teams could build new features without rebuilding data infrastructure
- Better User Experience: More relevant search results and intelligent recommendations
- Competitive Advantage: AI-powered features that differentiate Shopify from competitors
- Scalable Architecture: Systems that grow with the business
Conclusion: The Future of AI-Powered Commerce
The Product Understanding data assets we built are now utilized by numerous teams and contribute to many Shopify products. Our work demonstrates how strategic AI investments can transform entire product categories and create lasting competitive advantages.
The key insight? AI isn’t just about better algorithms—it’s about building systems that understand your business domain as well as your best employees do. When you get that right, the possibilities are endless.
References
As a technical leader with deep expertise in AI and machine learning, I’ve spent my career building systems that don’t just work—they transform how businesses operate. At Shopify, I led the team that built the AI foundation for intelligent e-commerce, demonstrating how strategic AI investments can create lasting competitive advantages.
Interested in learning more about AI strategy or technical leadership? Connect with me on LinkedIn or explore my other work in AI and machine learning.