Technology without strategy is doomed to fail. Looking at MLOps and Robotic Processes Automation exemplifies the need for clarity and strategic thinking.

Driving the AI strategy in one of the oldest French retailers is a real challenge in and of itself. Using AI to make educated guesses about which customers to target, what products to recommend, and where to display ads is even trickier. Aquarone explains how, as the company’s AI portfolio grew, a standardized canvas to go from a few dozen models to hundreds or even thousands of models became necessary. His team created a set of tools, guidelines, flows, and processes to drive this growth and make sure every prediction is understood, maintained, and used. He details the building blocks of the MLOps Platform and the milestones in transforming the organization to become AI-driven.

For too long, enterprises have lacked the capability to make use of unstructured data. While new technology is the obvious first tool that most organizations look to, savvy automation leaders know that the existing tech stack can be rife with opportunity to yield even more benefit. Wilde identifies ways that enterprises can leverage technology that integrates the existing tech stack to provide a richer, more comprehensive view of enterprise knowledge, while adding structure to their unstructured data. In turn, this strategy unlocks far greater value from existing automation solutions, such as Robotic Process Automation (RPA), and other business-critical technologies, illustrated through use cases in commercial real estate, financial services, insurance, and shared enterprise services.

Machine learning has evolved considerably over the past few years, and best practices have changed in tandem with that evolution.

On average, it takes 7 to 18 months, to go from idea to ML model in production. But things are changing. Data platforms are maturing and success is getting within reach of many organizations. Come hear about some top-level trends in machine learning.

Graph technology has become a main driver of AI and machine learning advances within a wide variety of industries.

The pandemic accelerated the pace of digital transformation across all industries. Organizations are looking for ways to accelerate their analytics, AI, and machine learning projects to increase revenue, manage risks, and improve customer experience. Join us to learn about the three core capabilities necessary to drive the business outcomes: connecting internal and external datasets and pipelines with a distributed graph database, analyzing connected data to discover insights with advanced analytics, and learning from the connected data with in-database machine learning.

Theories about machine learning have their place, but applying methodologies to real-world issues give practitioners a leg up.

Money laundering impacts society in a number of ways. Banks must adhere to regulatory guidelines to counter them. To combat money laundering, banks use data analytics to gain a complete understanding of transactions data accuracy, completeness, and timeliness from various sources within the bank. Maheshwari explains how extraction, transformation, and loading of data from various systems are critical. Machine learning methodologies are also critical to determine how these transactions (which are millions of dollars on a given day) can be analyzed for any money-laundering activity.

In this session, Hodeghatta addresses the challenges of protecting data while providing data for AI and machine learning projects and why data privacy is a concern and can be a hindrance for these types of projects.

Data within the enterprise is useless if it can’t be found and used. AI and ML provide some pathways to data discovery that give companies competitive advantage.

Machine learning is both a buzzword and the Holy Grail, depending on how you use it. Enterprise cloud companies use machine learning to accelerate or supercharge their data journey—it helps them work at a faster pace, with more efficiency and greater accuracy. Once AI is fully in use, teams need to be able to answer this question: How do we know if this is working correctly? In order to do this, teams must extend their existing observability approaches to cover their AI and ML capabilities. Today, there’s a big gap there, and that creates risk as organizations can’t manage those assets properly. This session takes a deep dive into how companies utilize machine learning to take every competitive advantage to advance their ability to use all the data at their disposal.

A single view of the customer, powered by AI and ML, can help identify fraud, personalize recommendations, and contribute to outstanding customer care encounters.

With the advent of omnichannel, leveraging customer data has become paramount. In a behemoth such as Walmart, each customer’s identity exists as a silo rather than a single customer from the company perspective. Handling customer data for privacy, regulations, and deprecation becomes challenging with every new ID introduced in the system. A streaming ML platform that seamlessly combines data belonging to the same customer on a single box and runs ML models, which use this data as features, leads to a single view of a customer. Brar discusses the platform (built on Kafka ecosystem) and its important aspects.

As the world becomes increasingly data-driven, AI/ML algorithms are being incorporated in most business applications.

Historically, data in AI architectures was moved to a central location to perform both model training and inference. This centralized approach is becoming untenable due to cost, performance, and privacy reasons. In this talk, Kaladhar shares his thoughts on next-generation distributed AI architectures and presents the concepts of “AI Marketplaces” and “Federated AI” to demonstrate how these concepts are an integral part of distributed AI architectures.

Putting AI to work to improve healthcare requires having technology and infrastructure coupled with the right people.

Healthcare innovators are innovators are prioritizing AI and operationalizing it for better performance, better outcomes, and better patient experience. But getting into predictive and prescriptive analytics takes many of them out of their comfort zones. Yet with every opportunity for improvement, there is a risk that organizations won’t have what it takes to successfully develop, implement, or operationalize AI. Having the technology and infrastructure for AI is not enough. Having the right people—with the technical capability and healthcare expertise—is enormously important. In this talk, Mehrotra and Fernando provide insight into strategies and plans to attract and retain the talent needed to unlock the potential of AI.