Editor’s note: In part one of this blog, Wayfair shared how it supports each of its 30 million active customers using machine learning (ML). Wayfair’s Vinay Narayana, Head of ML Engineering, Bas Geerdink, Lead ML Engineer, and Christian Rehm, Senior Machine Learning Engineer, take us on a deeper dive into the ways Wayfair’s data scientists are using Vertex AI to improve model productionization, serving, and operational readiness velocity. The authors would like to thank Hasan Khan, Principal Architect, Google for contributions to this blog.

When Google announced its Vertex AI platform in 2021, the timing coincided perfectly with our search for a comprehensive and reliable AI Platform. Although we’d been working on our migration to Google Cloud over the previous couple of years, we knew that our work wouldn’t be complete once we were in the cloud. We’d simply be ready to take one more step in our workload modernization efforts, and move away from deploying and serving our ML models using legacy infrastructure components that struggle with stability and operational overhead. This has been a crucial part of our journey towards MLOps excellence, in which Vertex AI has proved to be of great support.

Carving the path towards MLOps excellence

Our MLOps vision at Wayfair is to deliver tools that support the collaboration between our internal teams, and enable data scientists to access reliable data while automating data processing, model training, evaluation and validation. Data scientists need autonomy to productionize their models for batch or online serving, and to continuously monitor their data and models in production. Our aim with Vertex AI is to empower data scientists to productionize models and easily monitor and evolve them without depending on engineers. Vertex AI gives us the infrastructure to do this with tools for training, validating, and deploying ML models and pipelines.

Previously, our lack of a comprehensive AI platform resulted in every data science team having to build their own unique model productionization processes on legacy infrastructure components. We also lacked a centralized feature store, which could benefit all ML projects at Wayfair. With this in mind, we chose to focus our initial adoption of the Vertex AI platform on its Feature Store component.

An initial POC confirmed that data scientists can easily get features from the Feature Store for training models, and that it makes it very easy to serve the models for batch or online inference with a single line of code. The Feature Store also automatically manages performance for batch and online requests. These results encouraged us to evaluate the adoption of Vertex AI Pipelines next, as the existing tech for workflow orchestration at Wayfair slowed us down greatly. As it turns out, both of these services are fundamental to several models we build and serve at Wayfair today.

Empowering data scientists to focus on building world-class ML models

Since adopting Vertex AI Feature Store and AI Pipelines, we’ve added a couple of capabilities at Wayfair to significantly improve our user experience and lower the bar to entry for data scientists to leverage Vertex AI and all it has to offer:

1. Building a CI/CD and scheduling pipeline

Working with the Google team, we built an efficient CI/CD and scheduling pipeline based on the common tools and best practices at Wayfair and Google. This enables us to release Vertex AI Pipelines to our test and production environments, leveraging cloud-native services.

Keeping in mind that all our code is managed in GitHub Enterprise, we have dedicated repositories for Vertex AI Pipelines where the Kubeflow code and definitions of the Docker images are stored. If a change is pushed to a branch, a build starts in the Buildkite tool automatically. The build contains several steps, including unit and integration tests, code linting, documentation generation and automated deployment. The most important artifacts that are released at the end of the build are the Docker image and the compiled Kubeflow template. The Docker image is released to the Google Cloud Artifact Registry and we store the Kubeflow template in a dedicated Google Cloud Storage Bucket, fully versioned and secured. This way, all the components we need to run a Vertex AI Pipeline are available once we run a pipeline (manually or scheduled).

To schedule pipelines, we developed a dedicated Cloud Function that has the permissions to run the pipeline. This Function listens to a Pub/Sub topic where we can publish messages with a defined schema that indicates which pipeline to run with which parameters. These messages are published from a simple cron job that runs according to a set schedule on Google Kubernetes Engine. This way, we have a decoupled and secure environment for scheduling pipelines, using fully-supported and managed infrastructure.

Abstracting Vertex AI services with a shared library

We abstracted the relevant Vertex AI services currently in use with a thin shared Python library to support the teams that develop new software or migrate to Vertex AI. This library, called wf-vertex, contains helper methods, examples, and documentation for working with Vertex AI, as well as guidelines for Vertex AI Feature Store, Pipelines, and Artifact Registry.

One example is the run_pipeline method, which publishes a message with the correct schema to the Pub/Sub topic so that a Vertex AI pipeline is executed. When scheduling a pipeline, the developer only needs to call this method without having to worry about security or infrastructure configuration:

@cli.command()
def trigger_pipeline() -> None:
    from wf_vertex.pipelines.pipeline_runner import run_pipeline

    run_pipeline(
       template_bucket= f"wf-vertex-pipelines-{env}/{TEAM}",  # this is the location of the template, where the CI/CD has written the compiled templates to
       template_filename="sample_pipeline.json",  # this is the filename of the pipeline template to run
       parameter_values= {"import_date": today()}  # it’s possible to add pipeline parameters
)

Most notable is the establishment of a documented best practice for enabling hyperparameter tuning in Vertex AI Pipelines, which speeds up hyperparameter tuning times for our data scientists from two weeks to under one hour.

Because it is not yet possible to combine the outputs of parallel steps (components) in Kubeflow, we designed a mechanism to enable this. It entails defining parameters at runtime and executing the resulting steps in parallel via the Kubeflow parallel-for operator. Finally, we created a step to combine the results of these parallel steps and interpret the results. In turn, this mechanism allows us to select the best model in terms of accuracy from a set of candidates that are trained in parallel:

Our CI/CD, scheduling pipelines, and shared library have reduced the effort of model productionization from more than three months to about four weeks. As we continue to build the shared library, and as our team members continue to gain expertise in using Vertex AI, we expect to further reduce this time to two weeks by the end of 2022.

Looking forward to more MLOps capabilities

Looking ahead, our goal is to fully leverage all the Vertex AI features to continue modernizing our MLOps stack to a point where data scientists are fully autonomous from engineers for any of their model productionization efforts. Next on our radar are Vertex AI Model Registry and Vertex ML Metadata alongside making more use of AutoML capabilities. We’re experimenting with Vertex AI for AutoML models and endpoints to benefit some use cases at Wayfair next to the custom models that we’re currently serving in production.

We’re confident that our MLOps transformation will introduce several capabilities to our team, including: automated data and model monitoring steps to the pipeline, as well as metadata management, and architectural patterns in support of real-time models requiring access to Wayfair’s network. We also look forward to performing continuous training of models by fully automating the ML pipeline that allows us to achieve continuous integration, delivery, and deployment of model prediction services.

We’ll continue to collaborate and invest in building a robust Wayfair-focused Vertex AI shared library. The aim is to eventually migrate 100% of our batch models to Vertex AI. Great things to look forward to on our journey towards MLOps excellence.

When it comes to the adoption of artificial intelligence (AI), we have reached a tipping point. Technologies that were once accessible to only a few are now broadly available. This has led to an explosion in AI investment. However, according to research firm McKinsey, for AI to make a sizable contribution to a company’s bottom line, they “must scale the technology across the organization, infusing it in core business processes” — and based on conversations with our customers, we couldn’t agree more.

While investments in pure data science continue to be essential for many, widespread adoption of AI increasingly involves a category of applications and services that we call AI agents. These are technologies that let customers apply the best of AI to common business challenges, with limited technical expertise required by employees, and include Google Cloud products like Document AI and Contact Center AI. Today, at Google Cloud Next ‘22, we’re announcing new features to our existing AI agents and a brand new one with Translation Hub.

“AI is becoming a key investment for many companies’ long term success. However, most companies are still in the experimental phases with AI and haven’t fully put the technology into production because of long deployment timelines, IT staffing needs, and more,” said Ritu Jyoti, group vice president, worldwide AI and automation research practice global AI research lead, at IDC. “Organizations need AI products that can be immediately applied to automate processes and solve business problems. Google Cloud is answering this problem by providing fully managed, scalable AI agents that can be deployed fast and deliver immediate results.”

Translation Hub: An enterprise-scale translation AI agent

At I/O this year, we announced the addition of 24 new languages to Google Translate to allow consumers in more locations, especially those whose languages aren’t represented in most technology, to help reduce communication barriers through the power of translation. Businesses strive for the same goals, but unfortunately it is often out of reach due to the high costs that come with scaling translation.

That’s why today, we are announcing Translation Hub, our AI agent that provides customers with self-service document translation. With 135 languages, Translation Hub can create impactful, inclusive, and cost-effective global communications in a few clicks.

With Translation Hub, now researchers are able to share their findings instantly across the world, goods and services providers can reach underserved markets, and public sector administrators can reach more members of their communities in a language they understand — all of which ultimately help make for a more connected, inclusive world.

Translation Hub brings together Google Cloud AI technology, like Neural Machine Translation and AutoML, to help make it easy to ingest and translate content from the most common enterprise document types, including Google Docs and Slides, PDFs, and Microsoft Word. It not only preserves layouts and formatting, but also provides granular management controls such as support for post-editing human-in-the-loop feedback and document review.

“In just three months of using Translation Hub and AutoML translation models, we saw our translated page count go up by 700% and translation cost reduced by 90%,” said Murali Nathan, digital innovation and employee experience lead, at materials science company Avery Dennison. “Beyond numbers, Google’s enterprise translation technology is driving a feeling of inclusion among our employees. Every Avery Dennison employee has access to on-demand, general purpose, and company-specific translations. English language fluency is no longer a barrier, and our employees are beginning to broadly express themselves right in their native language.”

Document AI: A document processing AI agent to automate workflows

Every organization needs to process documents, understand their content, and make them available to the appropriate people. Whether it’s during procurement cycles involving invoices and receipts, contract processes to close deals, or for general increases in efficiency, Document AI simplifies and automates various document processing. With two new features launching today, Document AI can allow employees to focus on higher impact tasks and better serve their own customers.

For example, payments provider Libeo used Document AI to uptrain an invoice parser with 1,600 documents and increase its testing accuracy from 75.6% to 83.9%. “Thanks to uptraining, the Document AI results now beat the results of a competitor and will help Libeo save ~20% on the overall cost for model training over the long run,” said Libeo chief technology officer, Pierre-Antoine Glandier.

Today, we’re announcing these new features to our existing Document AI agent:

• Document AI Workbench can remove the barriers around building custom document parsers, helping organizations extract fields of interest that are specific to their business needs. Relative to more traditional development approaches, it requires less training data and offers a simple interface for both labeling data and one-click model training.
• Document AI Warehouse can eliminate the challenges that many enterprises face when tagging and extracting data in documents by bringing Google’s Search technologies to Document AI. This feature can make it simpler and easier to search for and manage documents like workflow controls to accommodate invoice processing, contracts, approvals, and custom workflows.

Contact Center AI: A contact center AI agent to improve customer experiences

Scaling call center support can be expensive and difficult, especially when implementing AI technologies to support representatives. Contact Center AI is an AI agent for virtually all contact center needs, from intelligently routing customers, to facilitating handoffs between virtual and human customer support representatives, to analyzing call center transcripts for trends and much more.

Just days ago, we announced that Contact Center AI Platform is now generally available to provide additional deployment choice and flexibility. With this addition to Contact Center AI, we are furthering our commitment to providing an AI agent that can assist organizations to quickly scale their contact centers to improve customer experiences and create value via data-driven decisions.

Dean Kontul, division chief information officer at KeyBank, had this to say about powering their contact center with Contact Center AI from Google Cloud: “With Google Cloud and Contact Center AI, we will quickly move our contact center to the Cloud, supporting both our customers and agents with industry-leading customer experience innovations, all while streamlining operations through more efficient customer care operations.”

Start delivering business results with AI agents, today!

If you’re ready to get started with Translation Hub, this Next ‘22 session has the details, including a deeper dive into Avery Dennison’s use of the AI agent.

To learn more about our Document AI announcements, check out our session with Commerzbank, “Improve document efficiency with AI,” as well as “Transform digital experiences with Google AI powered search and recommendations.”

And, to explore Contact Center AI Platform, watch “Delight customers in every interaction with Contact Center AI,” featuring more insight into KeyBank’s use case.