This is our 3rd Google Next conference (Read our 2017 Recap and 2018 Recap). Here are some of our observations and takeaways from the event.
One of the biggest changes last year was the announcement of Thomas Kurien as the new CEO of Google Cloud. Customers, Partners, and attendees including us were curious to learn about his vision, strategy and execution plans.
Real Customers
As we walked the expo halls and sessions, one of the biggest differences compared to the last 2 years is customers, real enterprise customers using Google Cloud in production. Enterprises from major verticals such as Retail, Healthcare, Energy, Financial Institutions and Public Sector shared their success stories on Google Cloud.
Product Announcements
Its well known that Google builds some great technology. During this year’s conference, there were several key product announcements. We are highlighting a few of our favorites:
Anthos
Google Cloud announced Anthos to solve a complex but very relevant for enterprises. All of the enterprises are looking to get started on AI but one of the biggest roadblocks is application modernization. Anthos is built on Kubernetes engine and provides deployment options for enterprises to choose hybrid cloud or any public cloud provider of choice.
Also support for open source platforms e.g Confluent, DataStax, Elastic, InfluxData, MongoDB, Neo4j and Redis. We have successfully migrated these platforms to Google Cloud to help Enterprises take advantage of the Google Cloud features.
We are excited about this announcement as we feel this will help expedite AI adoption.
BigQuery is a Platform
We all love BigQuery, such a fantastic product that simply works and has infinite scale. At SpringML, we track the success of an ML project by not on algorithm scores but how quickly the data is available in BigQuery. Everyone recognizes BigQuery as a data warehouse but the recent announcements indicate that it is much more.
One of the hard problems to address in ML projects is operationalizing the models. With BigQuery ML, We can create models as simple SQL queries! Similarly, AutoML Tables has integration to BigQuery that allows analysts to build predictive models without writing any code.
We recently implemented a project were we built Customer Segmentation and Life-Time Value models within BigQuery.
Salesforce-Google Cloud Partnership
This is an exciting announcement for us as we work closely with both Salesforce and Google Cloud. We believe Salesforce service cloud customers can take advantage of Google Cloud ML features such as Speech to Text, Language translation, and Dialogflow capabilities to improve the customer support experience.
Data Integration
Data access is one of the key topics that impact the success of any Analytics or ML implementation. Google Data Fusion allows building ETL pipelines right within Google Cloud environment. This will help customers migrate data and build a centralized data warehouse to support Analytics and ML Projects.
Customers are looking for Solutions
As highlighted by Thomas Kurien in his keynote, customers are looking for industry vertical solutions and not just technical services. At SpringML we have built solutions to address specific use cases. Here are a few solutions that we would like to highlight:
Retail
Instore Inventory Analysis. Eliminating tedious manual inventory checks with image analysis to immediately understand stock levels without human input. Here is a recent blog that highlights the solution.
Real-Time Product Search. End to end data pipeline to modernize the product search and improve the search results. Click here for a recent success story on the implementation we did for Kroger.
Healthcare
Intelligent Document Processing. Analyzing texts on clinical trial documents to extract key entities and sentiments. Click here to download the whitepaper.
Manufacturing
Automated Quality Control and Preventive Maintenance Using Visual Inspection. Click here to read a recent blog that details the solution.
Banking
Marketing Analytics. Digital marketing is an effective way to engage customers. However understanding the campaign performance helps to build the right media mix attribution, increase conversion and recommend the right products. SpringML has built a solution that addresses this use case. More details here.
Our Thoughts
Here at SpringML, our stance on Google Cloud as a long-term strategic partner remains the same: we are bullish and extremely interested to see what’s coming up in the future.
Google’s investments in sales and their partner ecosystem are a clear indication that they can not only build products but also are a serious contender in the enterprise space. While Google is undoubtedly playing catch-up in the public cloud, they are definitely making some serious moves that will set them up well for future success.
