New eBay platform using AI to enable image search and internal innovation – TechRepublic
Many of the biggest tech companies like Google, Facebook and Amazon have realized the value of creating their own AI platforms for both internal and customer-facing services.
Facebook’s FBLearner Flow helps the social media site filter out offensive posts, while Uber’s Michelangelo gives users time predictions for food deliveries.
To keep up with the competition, eBay has unveiled its AI platform, Krylov, which has given the company a wide range of new capabilities from improved language translation services to searching with images.
In a blog post, eBay’s Sanjeev Katariya, vice president and chief architect of the eBay AI and platforms, and Ashok Ramani, director of product management, computer vision, natural and language processing, discussed the creation of Krylov and how it has changed things both inside eBay and for users of the site.
“With computer vision powered by eBay’s modern AI platform, the technology helps you find items based on the click of your camera or an image. Users can go onto the eBay app and take a photo of what they are looking for and within milliseconds, the platform surfaces items that match the image,” Katariya and Ramani wrote in December.
“The user has not only activated computer vision technology, but they have also tapped into some advanced AI capabilities, including deep learning, distributed training and inferencing. The computer vision algorithm sifts through more than half a billion images and eBay’s 1.4 billion listings to find and show you the most relevant listings that are visually similar.”
SEE: [Artificial intelligence: A business leader’s guide (free PDF)] (TechRepublic Download)
The effort to create Krylov, which is named after famed mathematician Nicolai Krylov, evolved over time as eBay began to take stock of all its data and how it was innovating, they said.
According to Katariya and Ramani, data scientists at eBay now use Krylov to run thousands of model training experiments per month spanning a variety of AI use cases including computer vision, natural language processing, merchandising recommendations, buyer personalization, seller price guidance, risk, trust, shipping estimates and more.
Data scientists at eBay used to spend weeks and months building models to test new features for the website, wasting time, money and energy on processes that needed to be sped up and streamlined, they said.
Krylov has enabled eBay’s scientists to automate model training and deploy the models over individualized or a common inference as a platform in days instead of months.
The platform is now used to improve eBay’s recommendation system and enable powerful image search features that have opened up an entirely new avenue for the kind of searches users can perform. Shoppers can now simply upload a photo of what they want and browse for similar items.
The company says it has troves of data and Krylov allows eBay to harness the power of their information so they can evolve to better help their users.
With Krylov, eBay’s scientists have access to Notebooks, Tensorflow, PyTorch and H20 as well as the ability to train models like BERT or ResNet at scale.
“Krylov allows our AI teams to maximize the power of the vast repositories of data, both batch and real time, that eBay has. If you think of data as the fuel for artificial intelligence and machine learning, Krylov is the sophisticated vehicle being powered by that fuel,” Katariya and Ramani wrote.
They said the AI platform has been key for eBay’s machine translation technology, a significant contributor to enabling cross-border trade which makes up nearly 60% of eBay’s international revenue.
In an interview, Katariya said a Unified AI Initiative Core Team was created to spearhead the initiative, and the group included employees working on the AI platform as well as others working on hardware, networks, storage and data services.
The platform also got a huge amount of input from every part of eBay that could benefit from Krylov, including departments working on engineering in ads, computer vision, NLP, risk, trust and marketing.
“We had so many engineers and scientists across the company who needed help creating models and pushing out their models to production. We needed a complete closed loop on life cycle management of machine learning algorithms that was obvious. We needed a unified AI platform to really bring data scientists and engineers, modes and management experimentation all together,” Katariya said.
He added that the creation of Krylov brought scientists and engineers and platform builders together and was organized through core teams as well as machine learning fellowship programs.
Katariya said every member of the team worked to educate each other, share code and build it out with quality.
“If I look back on how we have progressed, I’m super proud of the collaboration, of the transparency, of the internal open source and how the training and education has gone above and beyond to build a really powerful platform that is global and deals with the scale of eBay,” he said.
“Krylov took a while to come of age but the objective was clear, that it was to ensure that our engineers and scientists across the globe, no matter where they were, were capable of accessing the right data at the right time, be it real time or batch-orientated data lakes or data warehouses or transactional data in a programmatic fashion.”
Katariya noted the immense amount of data eBay has on its hands, highlighting the website’s 1.4 billion listings, 190 markets and 183 million users.
Krylov is now powerful enough to serve as the backbone of eBay’s advertising, merchandising, recommendation systems and personalization systems.
Katariya explained that eBay’s computer vision is now built on top of Krylov, allowing for it to take in images and perform object recognition or object detection. This allows for its platform to combine image searching with textual searching to come up with cohesive ranking systems.
“Like any parent, I’m very proud of what we’ve been able to achieve with Krylov. It is quite a substantial advancement in building a platform of this nature cause its a polyglot, you can program in multiple different languages,” Katariya said.
“It’s very unique in how we built it. It’s holistic. The construction itself was built with diversity and diversification in mind. Which has resulted in an amazingly powerful platform.”
Cloud Insights Newsletter
Your go-to knowledge base for the latest about AWS, Microsoft Azure, Google Cloud Platform, Docker, SaaS, IaaS, cloud security, containers, the public cloud, the hybrid cloud, the industry cloud, and much more.
Delivered Mondays