Instacart has already revolutionized grocery shopping by enabling delivery of groceries bought online in as little as an hour. The crux of the company? Their fleet of shoppers—the workers who shop for Instacart’s customers in the brick-and-mortar stores, communicate with them in real time to resolve any issues, and bring the food to their doorsteps thousands of times every hour.
Jeremy Stanley explains how Instacart uses deep learning to enable the most efficient shoppers ever, putting the company at the top of the food chart in the on-demand economy. In the past, implementing deep learning has been difficult, given that retailers don’t have great planogram data and can’t track shoppers’ movements precisely within stores. Enter AI and deep learning, which has allowing Instacart to predict the sequence that shoppers pick items in specific store locations—in some cases leading to an in-store time savings of more than 10%. There are 125 million households in the US. If Instacart achieves just a 5% market share and those customers placed one order per week, then even if shoppers only saved one minute per order with a list-sorting algorithm, it would result in 618 years of shopping time saved every year. (For more details, see the company’s recent post, “Deep Learning with Emojis (Not Math).”)
Jeremy explores the significant advancements across AI, data science, and deep learning that give Instacart shoppers the tools they need to be most the efficient shoppers possible and discusses the bigger picture—how AI technologies are enabling the on-demand delivery economy and the implications that these advancements will have on the retail industry.
Jeremy Stanley is the vice president of data science at Instacart, where he works closely with data scientists who are integrated into product teams to drive growth and profitability through logistics, catalog, search, consumer, shopper, and partner applications. Previously, Jeremy was chief data scientist and EVP of engineering at Sailthru, a company building data-driven solutions for marketers to drive long-term customer engagement and optimize revenue opportunities, where he was responsible for the intelligence in the marketing personalization platform, led development, operations, database, and engineering support teams, and partnered with the CTO to drive innovation and stability while scaling; the CTO of Collective, where he led a team of product managers, engineers, and data scientists creating technology platforms that used machine learning and big data to address challenging multiscreen advertising problems; and the founder of Ernst & Young’s global markets analytics group, which analyzed the firm’s markets, financial, and personnel data to inform executive decision making. Jeremy’s background in data-driven technology products spans a decade consulting with numerous global financial services firms on predictive modeling applications as a leader in the customer analytics advisory practice at Ernst & Young.
©2017, O'Reilly Media, Inc. • (800) 889-8969 or (707) 827-7019 • Monday-Friday 7:30am-5pm PT • All trademarks and registered trademarks appearing on oreilly.com are the property of their respective owners. • firstname.lastname@example.org