Personal blog of Yzmir Ramirez

Nearly three years after it was first launched, Amazon Web Services’ SageMaker platform has gotten a significant upgrade in the form of new features, making it easier for developers to automate and scale each step of the process to build new automation and machine learning capabilities, the company said.

As machine learning moves into the mainstream, business units across organizations will find applications for automation, and AWS is trying to make the development of those bespoke applications easier for its customers.

“One of the best parts of having such a widely adopted service like SageMaker is that we get lots of customer suggestions which fuel our next set of deliverables,” said AWS vice president of machine learning, Swami Sivasubramanian. “Today, we are announcing a set of tools for Amazon SageMaker that makes it much easier for developers to build end-to-end machine learning pipelines to prepare, build, train, explain, inspect, monitor, debug and run custom machine learning models with greater visibility, explainability and automation at scale.”

Already companies like 3M, ADP, AstraZeneca, Avis, Bayer, Capital One, Cerner, Domino’s Pizza, Fidelity Investments, Lenovo, Lyft, T-Mobile and Thomson Reuters are using SageMaker tools in their own operations, according to AWS.

The company’s new products include Amazon SageMaker Data Wrangler, which the company said was providing a way to normalize data from disparate sources so the data is consistently easy to use. Data Wrangler can also ease the process of grouping disparate data sources into features to highlight certain types of data. The Data Wrangler tool contains more than 300 built-in data transformers that can help customers normalize, transform and combine features without having to write any code.

Amazon also unveiled the Feature Store, which allows customers to create repositories that make it easier to store, update, retrieve and share machine learning features for training and inference.

Another new tool that Amazon Web Services touted was Pipelines, its workflow management and automation toolkit. The Pipelines tech is designed to provide orchestration and automation features not dissimilar from traditional programming. Using pipelines, developers can define each step of an end-to-end machine learning workflow, the company said in a statement. Developers can use the tools to re-run an end-to-end workflow from SageMaker Studio using the same settings to get the same model every time, or they can re-run the workflow with new data to update their models.

To address the longstanding issues with data bias in artificial intelligence and machine learning models, Amazon launched SageMaker Clarify. First announced today, this tool allegedly provides bias detection across the machine learning workflow, so developers can build with an eye toward better transparency on how models were set up. There are open-source tools that can do these tests, Amazon acknowledged, but the tools are manual and require a lot of lifting from developers, according to the company.

Other products designed to simplify the machine learning application development process include SageMaker Debugger, which enables developers to train models faster by monitoring system resource utilization and alerting developers to potential bottlenecks; Distributed Training, which makes it possible to train large, complex, deep learning models faster than current approaches by automatically splitting data across multiple GPUs to accelerate training times; and SageMaker Edge Manager, a machine learning model management tool for edge devices, which allows developers to optimize, secure, monitor and manage models deployed on fleets of edge devices.

Last but not least, Amazon unveiled SageMaker JumpStart, which provides developers with a searchable interface to find algorithms and sample notebooks so they can get started on their machine learning journey. The company said it would give developers new to machine learning the option to select several pre-built machine learning solutions and deploy them into SageMaker environments.

Stacklet co-founders Travis Stanfield and Kapil Thangavelu met while both were working at Capital One several years ago. Thangavelu helped create the Cloud Custodian open-source cloud governance project. The two eventually got together and decided to build a startup based on that project and today the company launched out of stealth with a $4 million seed investment from Foundation Capital and Addition.

Stanfield, who is CEO at the young startup, says that Cloud Custodian came about as Capital One was moving to a fully cloud approach in around 2013. As the company looked for ways to deal with compliance and governance, it found that organizations like theirs were forced to do one-off scripts and they were looking for a way that could be repeatable and scale.

“Cloud Custodian was developed as a way of understanding what all those one-off scripts were doing, looking at the cloud control plane, finding the interesting set of resources, and then taking sensitive actions on them,” he explained.

After leaving Capital One, and going off in different directions for a time, the two came together this year to start Stacklet as a way to nurture the underlying open-source project Thangavelu helped build, and build a commercial company to add some functionality to make it easier for enterprises to implement and understand.

While cloud administrators can download and figure out how to use the raw open source, Stacklet is attempting to make that easier by providing an administrative layer to manage usage across thousands of cloud accounts along with pre-packaged sets of common kinds of compliance requirements out of the box, analytics to understand how the tool is doing and what it’s finding in terms of issues, and finally a resources database to understand all of the cloud resources under management.

The company has just three employees, including the two founders, but will be adding a couple of more shortly with a goal of having a team of 10 by year’s end. The open-source project has 270 contributors from around the world. The startup is looking to build diversity through being fully remote. Not being limited by geography means they can hire from anywhere, and that can help lead to a more diverse group of employees.

The founders admit that it’s a tough time to start a company and to be fundraising, but on the bright side, they didn’t have to be on a plane to San Francisco every week during the process.

In fact, Sid Trivedi, partner at Foundation Capital, said that this was his first investment where he never met the founders in person, but he said through long discussions he learned “their passion for the opportunity at hand, experience of the market dynamics and vision for how they would solve the problem of meeting the needs of both IT/security admins and developers.”

When you think of old, giant mainframes that sit in the basement of a giant corporation, still doing the same work they did 30 years ago, chances are you’re thinking about a financial institution. It’s the financial enterprises, though, that are often leading the charge in bringing new technologies and software development practices to their employees and customers. That’s in part because they are in a period of disruption that forces them to become more nimble. Often, this means leaving behind legacy technology and embracing the cloud.

At TC Sessions: Enterprise, which is happening on September 5 in San Francisco, Capital One executive VP in charge of its technology operations, George Brady, will talk about the company’s journey from legacy hardware and software to embracing the cloud and open source, all while working in a highly regulated industry. Indeed, Capital One was among the first companies to embrace the Facebook-led Open Compute project and it’s a member of the Cloud Native Computing Foundation. It’s this transformation at Capital One that Brady is leading.

At our event, Brady will join a number of other distinguished panelists to specifically talk about his company’s journey to the cloud. There, Capital One is using serverless compute, for example, to power its Credit Offers API using AWS’s Lambda service, as well as a number of other cloud technologies.

Before joining Capital One as its CTO in 2014, Brady ran Fidelity Investment’s global enterprise infrastructure team from 2009 to 2014 and served as Goldman Sachs’ head of global business applications infrastructure before that.

Currently, he leads cloud application and platform productization for Capital One. Part of that portfolio is Critical Stack, a secure container orchestration platform for the enterprise. Capital One’s goal with this work is to help companies across industries become more compliant, secure and cost-effective operating in the public cloud.

Early-bird tickets are still on sale for $249; grab yours today before we sell out.

Student tickets are just $75 — grab them here.

 If every company is truly a software company, Capital One is out to the prove it. It was one of the early users of Critical Stack, a tool designed to help build security into the container orchestration process. In fact, it liked it so much it bought the company in 2016, and today it’s releasing Critical Stack in Beta. This is a critical step toward becoming a commercial product, giving… Read More

 Snowflake, makers of a cloud data warehouse service, announced a new virtual private product that should appeal to highly regulated companies like financial services and healthcare. In fact, the company also announced that one of the product’s earliest customers, Capital One, will be investing $5 million in Snowflake as a strategic investor as a result of this new approach. Most… Read More

 Capital One is a huge organization with lots of compliance issues related to being a financial services company. It also happens to be an Amazon Web Services customer and it needed a tool to set rules and policies in an efficient way around AWS usage. Last July it started developing the tool that would become Cloud Custodian; today it announced at an AWS event in Chicago that it was making… Read More