![\"Retail](\"https:\/\/dmsretail.com\/RetailNews\/wp-content\/uploads\/2022\/05\/RETAIL-ONLINE-TRAINING-728-X-90.png\")
<\/a><\/p>\n<\/p>\n
How Instacart is working with NVIDIA to build a continuous learning intelligence layer for omnichannel retail<\/i><\/strong><\/p>\n For years, retailers have spoken about physical and digital commerce as separate channels. Instacart, a leading grocery technology company, believes that, in grocery, success in the future will be defined by how intelligently those worlds connect.<\/span><\/p>\n The goal is a continuous learning system that understands customers, baskets, shelves and store operations in real time, then using that intelligence to make shopping more personalised, efficient and profitable.<\/span><\/p>\n For David McIntosh, Instacart\u2019s chief connected stores officer, this is the idea behind Connected Store, the company\u2019s suite of enterprise technologies designed to digitise the in-store experience and link it with online commerce.<\/span><\/p>\n \u201cSince the very beginning, our enterprise business has been fundamental to Instacart,\u201d he says. \u201cWe power storefronts for hundreds of retailers in North America and growing globally.\u201d<\/span><\/p>\n But as those relationships deepened, retailers began asking Instacart for help beyond ecommerce. \u201cThey were saying, \u2018Instacart, you brought my business online, but I also have all of these problems in the store that I\u2019m dealing with. I haven\u2019t really digitised my store.\u2019\u201d<\/span><\/p>\n Instacart\u2019s ambition, he adds, is that \u201cin a five to 10 year horizon, customers won\u2019t have to choose between shopping in store or online. It will be one single unified mode.\u201d<\/span><\/p>\n That vision is compelling because grocery is so difficult to digitise well. Wi-Fi can be patchy, perhaps GPS breaks down indoors and even store lighting varies. Planograms are often incomplete or out of date. Shelves change constantly as customers shop, and staff replenish shelves and while brands rotate packaging. Even formats and seasonal ranges factor into a constantly-shifting environment.\u00a0<\/span><\/p>\n And, a single store can carry tens of thousands of SKUs. Recognition becomes difficult when shoppers place products upside down, cover labels, stack items together or move them in and out of a basket at speed.<\/span><\/p>\n As McIntosh puts it: \u201cThe retail environment is a business of exceptions.\u201d<\/span><\/p>\n That\u2019s exactly why Instacart has invested in Physical AI, or AI systems designed to understand and act on real-world environments. Instacart\u2019s Caper Carts AI smart trolleys that are fitted with basket-facing camera sensors, weights and measures-certified scales, location systems, outward-facing cameras and NVIDIA Jetson edge computing on every trolley.<\/span><\/p>\n Caper Carts look familiar because that\u2019s part of the point. \u201cWe believe smart carts are the winning format of retail precisely because it so nicely integrates in the store,\u201d says McIntosh. \u201cYou\u2019ve already shopped with a shopping trolley. That\u2019s a well-worn behaviour now for 80 years. It\u2019s the same way you shop, but it\u2019s better.\u201d<\/span><\/p>\n The screen on each Caper Cart gives customers a running total, deals, discounts and recommendations, while sensors help the trolley understand what is being added, removed and purchased. Instacart says Caper Carts are now deployed across more than 100 US cities, tripling year-on-year, and coming soon to Morrisons in the UK.<\/span><\/p>\n For customers, the appeal is not only speed. \u201cWhen I talk to customers in the store, the number one thing they tell me they love is the running total,\u201d says McIntosh. \u201cThe number two thing is deals and recommendations, which make the experience more personal.\u201d<\/span><\/p>\n Caper Carts use multimodal sensor fusion, combining signals from cameras, scales, location systems and shelf-facing views. An edge encoder running on NVIDIA Jetson provides real-time feedback, while cloud vision-language model encoders support deeper reasoning.<\/span><\/p>\n Cameras can triangulate where items are in 3D space, but they can be blocked. Weight can reveal what has changed, but it can drift, such as if the cart moves, it may take a few seconds to stabilise. McIntosh compares it to a diving board. After someone jumps, \u201cit takes time to stabilise. A lot can happen within three seconds.\u201d<\/span><\/p>\n By fusing these signals, Instacart aims to create an accurate view of the basket, the customer\u2019s location and the shelf environment. That matters because poor recommendations can do more harm than good.<\/span><\/p>\n \u201cIf you think the toothpaste is in one aisle, but in fact it\u2019s two aisles over, you\u2019re going to make the wrong recommendation,\u201d says McIntosh. \u201cThen customers are going to start to ignore the recommendations.\u201d<\/span><\/p>\n This is where the edge-to-cloud model becomes important. Stores cannot rely on constant connectivity or slow cloud round trips. \u201cOften, in a retail condition, you don\u2019t have Wi-Fi half the time in the store,\u201d McIntosh says. \u201cWhen you put that thing in the trolley, you need to know instantly, in hundreds of milliseconds.\u201d<\/span><\/p>\n The store as a measurable environment<\/b><\/p>\n The bigger prize is not just a smarter cart, but a more observable store.<\/span><\/p>\n Instacart captures millions of sensor inputs every day, from items entering and leaving baskets to cart movement, shelf conditions and screen engagement. Combined with more than 1.6 billion lifetime grocery orders, that creates what McIntosh describes as a continuously learning \u201cdata flywheel\u201d allowing partners to harness Instacart\u2019s grocery data expertise to understand and interpret their own data.<\/span><\/p>\n \u201cThe magic here is the purchase and location pair,\u201d he says. \u201cIt\u2019s \u2018where am I buying it in the store?\u2019 I\u2019m actually grabbing it off the shelf. This goes well beyond heat maps in terms of value.\u201d<\/span><\/p>\n For brands and retailers, that could mean understanding which displays drive sales, where products are bought, where gaps appear, and how assortment decisions should change by store. For store teams, it could eventually mean AI agents that flag shelves needing replenishment or identify out-of-stock risks.<\/span><\/p>\n \u201cYou start to make the store measurable,\u201d says McIntosh. \u201cIn some ways, you can start to make it programmable.\u201d<\/span><\/p>\n Retail media at the point of decision<\/b><\/p>\n The same intelligence also opens a potential retail media opportunity. In store, the Caper Cart screen can add context that online cannot by pulling in real-time data about store conditions and what\u2019s already in the trolley.<\/span><\/p>\n \u201cIf a cart already has one beverage, we could enable advertisers to market different brands, flavours, or types of beverages,\u201d says McIntosh. \u201cIt\u2019s totally native as you\u2019re in the store, as you\u2019re shopping.\u201d<\/span><\/p>\n The same intelligence also opens a potential retail media opportunity. In store, the Caper Cart screen can add context that online cannot by pulling in real-time data about store conditions and what\u2019s already in the trolley.<\/span><\/p>\n \u201cIf a cart already has one beverage, we could enable advertisers to market different brands, flavours, or types of beverages,\u201d says McIntosh. \u201cIt\u2019s totally native as you\u2019re in the store, as you\u2019re shopping.\u201d<\/span><\/p>\n Instacart\u2019s intelligence layer<\/b><\/p>\n Instacart\u2019s long-term vision is centered around its intelligence layer. A foundational AI system that understands not just products, but how grocery retail works as a living ecosystem. That means connecting product relationships, household preferences, store conditions, inventory, logistics, retail media and operational decision-making across the retailer\u2019s omnichannel data.<\/span><\/p>\n NVIDIA\u2019s role is central. NVIDIA Jetson runs at the edge on Caper Carts, Dynamo-Triton supports cloud ranking and personalisation, and Instacart expects to roll out Nemotron to help power agentic platforms.<\/span><\/p>\n The future of grocery may not be online or in store \u2013 it\u2019s a unified experience connected by an intelligence layer.<\/span><\/p>\n![\"\"](\"https:\/\/www.retailgazette.co.uk\/wp-content\/uploads\/2026\/06\/Screenshot-2026-06-17-at-10.23.29-am-300x224.png\")
A business of exceptions<\/b><\/p>\n![\"\"](\"https:\/\/www.retailgazette.co.uk\/wp-content\/uploads\/2026\/06\/Screenshot-2026-06-17-at-10.24.32-am-300x227.png\")
Accuracy that earns trust<\/b><\/p>\n