Technologies that use data to unlock value at the most natural points of human contact are maturing faster than ever — and this evolution will enable connected experiences across channels, said Will Hayes, Lucidworks CEO. He was speaking at Activate, our annual search and AI conference, held in Washington, D.C.this year.
Hayes opened the conference joking about the never-ending argument of whether we call this set of technologies machine learning or artificial intelligence. He said that no matter how we slice the cake or overlap the Venn, it’s a fool’s errand. Instead, he says, the focus should be on using these technologies to unlock user intention and leveraging data to delight users at every moment of the user experience.
Search remains the best way to unlock value from the world’s data. And the search interface is the easiest paradigm for employees and consumers to get the insights they need to do the things they want.
Many sessions this year focused on separating the sci-fi fever dreams of Skynet and robots taking over the world from the practical everyday data-driven applications of machine learning that we are starting to see in production for both consumer and business applications. This illustrated the difference between pure AI that approximates the breadth and depth of human intelligence versus practical AI which focuses on specific tasks, use cases, or industries.
Data Experience is the New User Experience
Hayes talked about a four-part maturity model showing how companies are evolving their search and discovery capabilities and what that means specifically for a few brands.
He illustrated this with the evolution in ancient cartography from going to AAA for your summer vacation TripTik to summoning up directions on MapQuest to Uber’s app that makes getting from point A to B as simple as the push of a button. Consumers expect this ease in the apps and services they use in their everyday lives – and increasingly in the workplace
He described how Lululemon is using AI and machine learning to understand the intention of shoppers to make a seamless experience from physical store to phone support to online store. Making returns and refunds is as easy as walking into a local store to exchange sizes or get a refund.
Hayes also referenced his own childhood love of Mad magazine and sharing issues and clippings with friends and how that filters down into how Reddit has become the space for sharing and enjoying the entire Internet in one single, delightful user experience.
All of these put the user research and understanding users at the forefront in creating a powerful data experience.
Along with the greater discussion about search, AI, and the future, Hayes also talked about some brand new Lucidworks product announcements. Lucidworks continues to bring open source technologies such as Apache Solr and Spark to the enterprise to transform the way people do their jobs, and there were a few announcements in line with that mission.
Simplifying DevOps with Lucidworks Managed Search
Another big announcement from the opening keynote focused on the practical, every day tasks and trials of managing a Solr cluster. For all of its power and reliability, running your Solr infrastructure isn’t always as straightforward as it should be. Teams have to pull their Solr experts off development time, creating and improving applications, to focus on cumbersome devops tasks such as uptime, upgrades, and scalability.
To remedy this problem, Hayes announced the availability of Lucidworks Managed Search, a hosted SaaS offering providing cost effective and simplified operations for the most reliable and trusted open source search technology. Lucidworks Manages Search helps autoscale and maintain high availability of their Solr clusters. LMS also includes security, push-button deployments, and a variety of other developer tools and integrations are available. Deploying and Managing Apache Solr has never been easier.
Fusion 5 Launches with Python Support
And of course, like in past years, Hayes unveiled new features and capabilities to the core platform Lucidworks Fusion, now out in version 5. He talked about how Fusion 5 includes native support for Python machine learning models, so data science teams can plug their custom machine learning models directly into Fusion and integrate with their existing data science infrastructure and tools. This allows data science teams to augment user’s human intelligence across the entire experience. This enables applications for question and answering, chat bots, image recognition and much more.
But the big news for Fusion 5 is the entire platform has been re-architected to be cloud native and run on a microservices infrastructure. Fusion 5’s cloud native architecture breaks Fusion down to a series of microservices, which can be independently deployed, scaled and upgraded. By delivering these capabilities in a cloud native format, we are increasing the speed and value you get in bringing data driven Applications to market.
Optimize the Shopping Experience with Predictive Merchandiser
Another new innovation is the Predictive Merchandiser which lets customers optimize the shopping experience with drag-and-drop ease This frees merchandisers from the repetitive grunt work of having to go to IT to implement and change business rules. Predictive Merchandiser gives our customers the power to create the best shopping experience possible. This way those who understand their domain and product categories best and can make the most impact. And we’re letting merchandisers focus their creativity on improving conversions, increasing revenue, and delighting shoppers.
Full details about the Predictive Merchandiser add-on for Lucidworks Fusion.
A Burgeoning Community and Industry
Hayes closed with some quick stats about this year’s Activate conference made up of over 1,000 search and AI professionals from all over the world. We also looked at stats showing growth in the overall space gauged by demand for data scientists, search engineers, and data engineers.
…matched against growth in career potential…
…and growth in the global community of open source, Solr, and AI practitioners.
Closing The Final Mile
The closing thoughts of Hayes’s keynote were that these technologies are maturing faster than ever, and this evolution will enable connected experiences across channels. That by Solving this “final mile” problem means that we can treat applications and channels, not as siloed experiences, but as sources of insight for each and every user.