Business Problem
UKG Ready primarily operates in the Small and Medium Business (SMB) space, so inherently many customers are forced to operate and make key business decisions with less Workforce Management (WFM) / Human Capital Management (HCM) data. In addition to volume, SMB lacks the variety of data needed to create a dynamic and agile organization. This puts SMB at a major disadvantage compared to larger segments.

Project Goals
People Insights module is committed to surfacing insights to customers in the context of their day-to-day duties and aid in decision making. With the SMB customer data limitations mentioned above, the goal of this project was to create a global dataset that augments individual customer data to bring light to less obvious, yet important information.

Challenges
UKG Ready is a highly configurable application that gives customers the opportunity to build solutions on a platform that meets their specific business needs. High configurability gives high flexibility to customers in their usage of the software. However, it becomes nearly impossible to create a global dataset for machine learning and data insights. UKG Ready manages just under 4 million of the US workforce and some 30,000+ customers. Despite the large employee dataset size, machine learning models that are specific to customers are starved for data because the individual customers have a relatively small employee population. Does that mean we cannot support our SMB customers’ decision making with ML?

Result
Partnering with Google, we were able to develop an approach that allowed us to standardize various domain entities (pay categories, time off codes, job titles, etc.) so that we could build a global dataset to augment SMB customer data. Using machine learning we were able to build a common vocabulary across our customer base. This common vocabulary encapsulates the nuances of how our customers manage their business and yet is generalized and standardized such that the data can be aggregated over the variety of customer configurations. This allows us to serve up practical insights to customers through various use cases. Our partnership allowed us to leverage Google Cloud Services to meet the needs of our complex machine learning models, distributed data sets and CI/CD processes.

How
UKG Ready decided to partner with Google for an end-to-end solution for the analytics offering. This allowed us to focus on our core business logic without having to worry about the platform, environment configurations, performance and scalability of the entire solution. We make use of various Google Cloud services such as Cloud Triggers, Cloud Storage, Cloud Functions, Cloud Composer, Cloud Dataflow, Big Query, Vertex AI, Cloud Pub/Sub… to host our analytics solution. Jenkins manages the entire CI/CD pipelines and cloud environments are configured and deployed using Terraform.

The standardization of business entities problem was solved in three distinct steps:

Step 1: Collecting aggregated data
We needed an approach to collect aggregated data from our highly distributed, sharded, multi-tenant data sources. We developed a custom solution that allows us to extract data aggregated at source for PII and GDPR considerations and transfer to Google Cloud Storage in the fastest manner possible. Data is then transformed and stored in Big Query. Services used: GCS, Cloud Functions, DataFlow, Cloud Composer and Big Query. All processes are orchestrated using Cloud Composer and detailed logging is available in Cloud Logging (Stackdriver).

Step 2: Applying NLP (Natural Language Processing)
Once we had the variety of customer configurations or the business entities available, we then applied NLP algorithms to categorize and standardize these in buckets. This approach assumes that customers use natural language for configurations like job titles, pay codes etc.

String Preparation
The input data for string preparation process is an entity string or several strings, that describe one entity object (like name-description pair or code-name pair). The output represents set of tokens that may be used to run a classification/clustering model. The process of string preparation tokenizes strings, replaces shortcuts, handles abbreviations, translates tokens, handles grammatical errors and mistypes

ML Models

Statistical
The idea of the model is to use defined target classes (clusters) and assign several tokens (anchors) to each of them an entity that has any of those tokens would be “attracted” to appropriate class. All other tokens are weighted according to frequencies of usage of theses tokens in the entities with anchor tokens:

Using anchor tokens, we are building kind-of Word2Vec - dimensionality of vector is equal to number of target classes. The higher the specific dimension (cluster) value, the higher the probability of entity to be included in appropriate cluster. Final prediction for entity tokens list for specific class is sum of weights of all the tokens included. Predicted cluster is a cluster that has maximal prediction score.

Lexical Model
We managed to generate reasonable amount of labeled data during statistical model implementation and testing. That opens a possibility to build “classical” NLP model that uses labeled data to train classification neural network using pretrained layers to produce token embeddings or even string embeddings. We started experimentation with pre-trained models like GloVe and got good results with single words and bi-grams but started getting issues in handling of n-grams. Our Google account team came to our rescue and recommended some white papers that helped formulate our strategy. We now use Tensorflow nnlm-en-dim128 model to produce string embeddings – it was trained on 200B records English Google News corpus and produces for each input string 128-dimensional vector. After that we use several Dense and Dropout layers to build a classification model.

Ensembling
To perform ensembling all the model results for each class are cast to probabilities using softmax transformation with scale normalization. Final predicted probability is maximal average score of both models among all the classes scores – appropriate class is predicted class.

The machine learning models are deployed on Vertex AI and are used in batch predictions. Model performance is captured at every prediction boundary and monitored for quality in production.

Step 3: Making available common vocabulary
Having the standardized vocabulary, we then needed a mechanism to have the results be available in UKG Ready reports and customer specific models like Flight Risk and Fatigue. For this we again used Google Services for orchestration, data transformation and data storage.

Once the modeling is complete, we made the customer specific models leveraging the above architecture be available in Reports. We utilized our proven existing technology choices in GCP for orchestration, data transformation and data storage

Results
We are able to build a common vocabulary of our customers’ business entities with good confidence. And be an expert advisor to our SMB customers in their decision-making using machine learning. With the advice of our Google account team and using Google services we can add value to our product in a relatively short amount of time. And we are not done! We continue to use this platform for new use cases, complex business problems and innovative machine learning solutions.

Sample result:

Special thanks to Kanchana Patlolla , AI Specialist, Google for the collaboration in bringing this to light

Some of the most important data at your company isn’t living in databases, but in documents, and most business processes begin, involve or end with a document. 

Yet most companies are still manually entering data and reliant on guesswork to make sense of it all as the volume and variety of data explodes. Organizations are also leaving heaps of value on the table in the form of new and better customer experiences that can be unlocked with artificial intelligence (AI) applied to documents. 

The latest releases of Document (Doc) AI platform, Lending DocAI and Procurement DocAI, built on decades of AI innovation at Google, bring powerful and useful solutions to these challenges. Under the hood are Google’s industry-leading technologies:

• Computer vision (including OCR) and Natural Language Processing (NLP) that creates pre-trained models for high-value, high-volume documents. 
• Google Knowledge Graph to validate and enhance the fields in your documents.
• Training and creation of your own custom document models. 
• Human interaction with AI to ensure accuracy where needed.

Google Cloud DocAI platform, Lending DocAI and Procurement DocAI are now generally available. Thousands of customers have tried these products in the preview phase—and DocAI has already processed tens of billions of pages of documents across lending, insurance, government and other industries.

“The vast majority of enterprise content still resides in unstructured sources like documents. Google Cloud’s Document AI brings a fresh new perspective to the problem informed by the company’s decades of experience making sense of the largest unstructured corpus in the world—the world wide web.” —Ritu Jyoti, VP of AI Research, IDC

Cut document processing costs by up to 60%

Lending DocAI helps banks, mortgage brokers and other lending institutions fast track the loan application process from weeks to days, dramatically reducing the cost of issuing a loan. And Procurement DocAI enables companies to automate procurement data capture at scale, lowering processing costs by up to 60%.

These solutions are built on DocAI platform, a unified console for document processing that lets you quickly access all parsers and tools. From the platform, you can automate and validate documents to streamline workflows, reduce guesswork, and keep data accurate and compliant. 

Get more value from AI with DocAI’s industry-specific solutions

According to Accenture’s AI: Built to Scale report: “Companies that scale successfully see 3x the return on their AI investments compared to those who have not fully rolled out AI capabilities.” 

Core to our strategy at Google Cloud is the creation of industry-specific solutions that help companies get maximum value out of their investments in AI. We announced Lending DocAI, our first solution designed specifically for the financial services industry, at the Mortgage Bankers Association convention last year. It processes borrowers’ income and asset documents using a set of specialized machine learning (ML) models, and automates routine document reviews so that mortgage providers can focus on more important work. 

Lending DocAI is now generally available and includes more specialized parsers for critical loan documents including paystubs, bank statements, and more. Our goal is to provide the right tools to help borrowers and lenders have a better experience and close home loans faster. For more, watch this video.

Procurement DocAI is also now generally available. This solution helps companies accelerate document processing for invoices, receipts, and other valuable documents in the procurement cycle. 

Automating data capture is helping our customers increase accuracy and also lower their procure-to-pay processing costs. We are continually expanding the types of documents Procurement DocAI can process—the latest is a utility parser for electric, water and other bills. In addition, Procurement DocAI leverages Google Knowledge Graph to validate and enrich parsed information to make the data even more useful. Check out this overview video for more details. 

One company that lives and breathes AI-enabled document management is AODocs. It uses Procurement DocAI to simplify invoice processing for enterprise customers and launched a new Gmail add-on, Invoice to Sheet, for SMB customers who just want to track their invoices in Google Sheets.

“Google Cloud’s Procurement DocAI service allows our document management platform to better automate the processing of invoices; AODocs customers who have tested our new account payables workflow estimate that the productivity of their A/P team has more than doubled, thanks to the reduction of manual data input brought by the Procurement DocAI.”—Stéphan Donzé, Founder and CEO, AODocs

The new specialized parsers for Lending and Procurement DocAI can be used alongside our existing AutoML Text & Document Classification and AutoML Document Extraction services. These technologies provide a state-of-the-art toolset for creating new document models and have been widely deployed by customers in financial services and other industries. 

Partner to accelerate your AI deployment and results

Having the right partner to ease the complexity of rolling out your AI-strategy in mortgage document processing is critical to transforming your customers’ experience. We’re excited to announce a partnership with Mr. Cooper, a leader in mortgage servicing, to provide customers with more automation and workflow tools throughout their entire mortgage life cycle. As part of this agreement, both companies will collaborate on digitizing Mr. Cooper’s core mortgage platform, creating a more personal customer experience utilizing AI, and driving a broader culture of innovation to imagine and develop services and solutions that will transform the mortgage experience for American homeowners.

“Over the last few years, we have made substantial investments in our servicing technology and core mortgage platform that have revolutionized the customer experience, while providing dramatic efficiencies in operating cost. Our partnership with Google Cloud AI will build on those advances and help make these technologies available for the mortgage industry.” —Jay Bray, Chairman and CEO, Mr. Cooper Group

This builds upon the robust partner ecosystem we’re creating to help customers revolutionize the home loan experience, which includes last year’s partnership announcement with Roostify.

Integrate human review into ML predictions 

Next up is the general availability of Human-in-the-Loop AI, a new DocAI feature that will help companies achieve higher document processing accuracy with the assurance of human review. Adding human review can increase accuracy and help businesses interpret predictions using purpose-built tools to enable those reviews. 

Processing documents quickly and cost-effectively is important. But it’s often necessary to have a high level of assurance on data accuracy for compliance. CIOs and IT decision-makers need highly accurate ML predictions to fulfill compliance requirements, improve employee experience (e.g. less rework), and raise customer satisfaction (e.g. fewer data errors). Including human participation in ML processes allows AI and humans to work together for the best possible results.

Human-in-the-Loop AI provides the workflow to manage human review tasks and produces a percentage confidence score of how “sure” it is that the AI ingested the document correctly. Document AI extracts data from documents with ML, and when paired with Human-in-the-Loop AI, human reviewers are able to verify the data captured. This system is customizable, providing the flexibility to set different thresholds and assign individual groups of reviewers to various stages of the workflow. With Human-in-the-Loop AI, developers can choose trusted reviewers to assign to the task; these reviewers can be from within their own or partner organizations.

More Document AI resources

To learn more, check out the Document AI webpage and watch a demo of how to process sample forms in AI Platform notebooks to inspect data extraction and confidence scores. For more on how customers and partners like Workday, AODocs, and Mr. Cooper are using Document AI, listen to our fireside chat. And stay tuned for the exciting evolution of these technologies in future releases of DocAI.

First, What is Google Cloud Talent Solution?

Cloud Talent Solution is a service that brings machine learning to the job search experience, returning high quality results to job seekers far beyond the limitations of typical keyword-based methods. Once integrated with your job content, Cloud Talent Solution automatically detects and infers various kinds of data, such as related titles, seniority, and industry.

Show Me How it Works

Try it online now.

Show Me an Example of Who’s Using It

There’s a number of enterprises leveraging this service. Here are a few easy-to-watch examples

Watch how FedEx Ground Employs Google Cloud Talent Solution

Read how Johnson & Johnson is Reimagining Recruiting with Jibe and Google

How Much Does it Cost?

Ok, Let’s See How it Works

New AI startup program benefits and Accelerator introduced at the Google Cloud Startup Summit

Google Cloud is committed to supporting the growth and advancement of startups, with particular focus on helping startups looking to build and scale. We’re seeing tremendous innovation from startups choosing Google Cloud to advance their generative AI development, thanks to our fully-managed and serverless data, AI, and infrastructure solutions. That’s why today’s Startup Summit, and announcements, are focused on AI.

Following our March 14th announcement that we were expanding the Google for Startups Cloud Program with exclusive benefits for AI startups, we’re excited to announce that the program is now live. Starting today, eligible seed to series A startups that use AI as their core technology to develop their primary products or solutions can apply here. This gives them access to all the Google for Startups Cloud Program benefits, including:

Up to $350,000 USD over two years in Google Cloud credits 
For Google Cloud and Firebase usage covered in:

• Year 1: 100% up to $250,000 USD in Google Cloud credits [1] (includes standard program credits plus an additional $150,000 USD for the AI startup program)
• Year 2: 20% up to an additional $100,000 USD in Google Cloud credits

Technical & collaboration support
We’re providing credits to allow AI startups access to fast, high-quality Customer Care Enhanced Support, access to Google Cloud Startup Customer Engineers, and a dedicated Startup Success Manager to accelerate onboarding with Google Cloud. We’re also offering 12 months of free Google Workspace Business Plus for new sign-ups.

Additional AI benefits, such as access to AI experts, training, and resources include: 

Webinars & live Q&A sessions 
Exclusive access to join webinars and live Q&As with our Google Cloud AI product managers, engineers, and developers.

Insight into AI innovation
Direct visibility into Google Cloud’s latest AI advances and product roadmap.

Hands-on AI learning labs
Free access to advanced hands-on learning labs focused on AI/ML and the latest Google Cloud technology.

Exclusive AI ISV/Saas startup access to Built with Google Cloud AI
Access to the AI Center of Excellence, tools and training, co-marketing, and amplification on Google Cloud Marketplace. These resources have historically been available only to AI enterprise ISV/SaaS companies, but will now also be offered to ISV/SaaS startups in our program.

Dedicated AI technical guidance, workshops, and best practices
Architectural guidance and best practices to get started  and build with Google Cloud AI solutions, including technical deep dives and hands-on workshops.

We’re also excited to announce an inaugural North American Google for Startups Accelerator: Cloud. This 10-week virtual Accelerator is equity-free and best suited for cloud-native startups leveraging AI and ML in their operations.  Designed to help prepare for the next phase of the growth journey, participating startups will work with the best of Google’s programs, products, people, and technology. Applications are open now until May 30th and the program kicks off in July.  Find out more here. 

Our goal is to enable more AI-first startups and give them the technology, community, and resources they need to build and grow their startup faster, smarter, and cheaper.  Furthermore, partnering with Google Cloud lets startups seamlessly integrate with other Google solutions and leverage our global infrastructure. Google Cloud’s AI platform, Vertex AI, gives startups the intelligence they need to make smart business decisions and allows them to easily build, deploy, and scale machine learning (ML) models faster. They can also quickly build with Google Cloud’s advanced generative AI technologies or use our best-in-class speech, vision, translation, and language APIs. To learn more about Google Cloud’s recently-announced Generative AI Support in Vertex AI, check out this deep dive, and to keep up with Google Cloud’s latest generative AI news and thought leadership, read The Prompt on Transform with Google Cloud. 

Startups around the world are choosing Google Cloud. Join us and let’s build the future, together.

Establishing a mature MLOps practice to build and operationalize ML systems can take years to get right.  We recently published our MLOps framework to help organizations come up to speed faster in this important domain.  

As you start your MLOps journey, you might not need to implement all of these processes and capabilities. Some will have a higher priority than others, depending on the type of workload and business value that they create for you, balanced against the cost of building or buying processes or capabilities. 

To help ML practitioners translate the framework into actionable steps, this blog post highlights some of the factors that influence where to begin, based on our experience in working with customers. 

The following table shows the recommended capabilities (indicated by check marks) based on the characteristics of your use case, but remember that each use case is unique and might have exceptions. (For definitions of the capabilities, see the MLOps framework.)

Your use case might have multiple characteristics. For example, consider a recommender system that’s retrained frequently and that serves batch predictions. In that case, you need the data processing, model training, model evaluation, ML pipelines, model registry, and metadata and artifact tracking capabilities for frequent retraining. You also need a model serving capability for batch serving.

In the following sections, we provide details about each of the characteristics and the capabilities that we recommend for them.

Pilot

Example: A research project for experimenting with a new natural language model for sentiment analysis.

For testing a proof of concept, your focus is typically on data preparation, feature engineering, model prototyping, and validation. You perform these tasks using the experimentation and data processing capabilities. Data scientists want to set up experiments quickly and easily and track and compare them. Therefore, you need the ML metadata and artifact tracking capability in order to debug, to provide traceability and lineage, to share and track experimentation configurations, and to manage ML artifacts. For large-scale pilots, you might also require dedicated model training and evaluation capabilities.

Mission-critical

Example: An equities trading model where model performance degradation in production can put millions of dollars at stake.

In a mission-critical use case, failure with the training process or production model has a significant negative impact on the business (a legal, ethical, reputational, or financial risk). The model evaluation capability is important to identify bias and fairness, as well as to provide explainability of the model. Additionally, monitoring is essential to assess the quality of the model during training and to assess how it performs in production. Online experimentation lets you test newly trained models against the one in production using a controlled environment before you replace the deployed model. Such use cases also need a robust model governance process to store, evaluate, check, release, and report on models and to protect against risks. You can enable model governance by using the model registry and metadata and artifact tracking capabilities. Additionally, datasets and feature repositories provide you with high-quality data assets that are consistent and versioned.

Reusable and collaborative

Example: Customer Analytic Record (CAR) features that are used across various propensity modeling use cases.

Reusable and collaborative assets allow your organization to share, discover, and reuse AI data, source code, and artifacts. A feature store helps you standardize the processes of registering, storing, and accessing features for training and serving ML models. Once features are curated and stored, they can be discovered and reused by multiple data science teams. Having a feature store helps you avoid reengineering features that already exist, and saves time on experimentation. You can also use tools to unify data annotation and categorization. Finally, by using ML metadata and artifacts tracking, you help provide consistency, testability, security and repeatability of the ML workflows. 

Ad hoc retraining

Example: An object detection model to detect various car parts, which needs to be retrained only when new parts are introduced. 

In ad hoc retraining, models are fairly static and you do not retrain them except when the model performance degrades. In these cases, you need data processing, model training, and model evaluation capabilities to train the models. Additionally, because your models are not updated for long periods, you need model monitoring. Model monitoring detects data skews, including schema anomalies, as well as data and concept drifts and shifts. Monitoring also lets you continuously evaluate your model performance, and it alerts you when performance decreases or when data issues are detected. 

Frequent retraining

Example: A fraud detection model that’s trained daily in order to capture recent fraud patterns.

Use cases for frequent retraining are ones where model performance relies on changes in the training data. The retraining might be based on time intervals (for example, daily or weekly), or it could be triggered based on events like when new training data becomes available. For this scenario, you need ML pipelines to connect multiple steps like data extraction, preprocessing, and model training. You also need the model evaluation capability to ensure that the accuracy of the newly trained model meets your business requirements. As the number of models you train grows, both a model registry and metadata and artifact tracking help you keep track of the training jobs and model versions.

Frequent implementation updates

Example: A promotion model with frequent changes to the architecture to maximize conversion rate.

Frequent implementation updates involve changes to the training process itself. That might mean switching to a different ML framework, such as changing the model architecture (for example, LSTM to Attention) or adding a data transformation step in your training pipeline. Such changes in the foundation of your ML workflow require controls to ensure that the new code is functional and that the new model matches or outperforms the previous one. Additionally, the CI/CD process accelerates the time from ML experimentation to production, as well as reducing the possibility for human error. Because the changes are significant, online experimentation is necessary to ensure that the new release is performing as expected. You also need other capabilities such as experimentation, model evaluation, model registry, and metadata and artifact tracking to help you operationalize and track your implementation updates. 

Batch serving

Example: A model that serves weekly recommendations to a user who has just signed up for a video-streaming service.

For batch predictions, there is no need to score in real time. You precompute the scores and you store them for later consumption, so latency is less of a concern than in online serving. However, because you process a large amount of data at a time, throughput is important. Often batch serving is a step in a larger ETL workflow that extracts, pre-processes, scores, and stores data. Therefore, you need the data processing capability and ML pipelines for orchestration. In addition, a model registry can provide your batch serving process with the latest validated model to use for scoring.

Online serving

Example: A RESTful microservice that uses a model to translate text between multiple languages. 

Online inference requires tooling and systems in order to meet latency requirements. The system often needs to retrieve features, to perform inference, and then to return the results according to your serving configurations. A feature repository lets you retrieve features in near real time, and model serving allows you to easily deploy models as an endpoint. Additionally, online experiments help you test new models with a small sample of the serving traffic before you roll the model out to production (for example, by performing A/B testing).

Get started with MLOps using Vertex AI

We recently announced Vertex AI, our unified machine learning platform that helps you implement MLOps to efficiently build and manage ML projects throughout the development lifecycle. You can get started using the following resources: 

• MLOps: Continuous delivery and automation pipelines in machine learning
• Getting started with Vertex AI
• Best practices for implementing machine learning on Google Cloud

^{Acknowledgements: I’d like to thank all the subject matter experts who contributed,  including Alessio Bagnaresi, Alexander Del Toro, Alexander Shires, Erin Kiernan, Erwin Huizenga, Hamsa Buvaraghan, Jo Maitland, Ivan Nardini, Michael Menzel, Nate Keating, Nathan Faggian, Nitin Aggarwal, Olivia Burgess, Satish Iyer, Tuba Islam, and Turan Bulmus. A special thanks to the team that helped create this, Donna Schut, Khalid Salama, and Lara Suzuki, and Mike Pope for his ongoing support.}