Whitepaper
4 min read

eBook: Power Up Machine Learning Process. Build Feature Stores Faster - an Introduction to Vertex AI, Snowflake and dbt Cloud

Recently we published the first ebook in the area of MLOps: "Power Up Machine Learning Process. Build Feature Stores Faster - an Introduction to Vertex AI, Snowflake and dbt Cloud". In this short article, we will tell you what you can find inside the ebook, what questions and problems it addresses and share the first opinions. 

The Business Perspective and Technical Perspective on Feature Store and MLOps 

In this ebook, we discuss in detail the Feature Store as one of the components of MLOps. From the general concept of MLOps and Feature Store as a solution, to specific problems that cause inefficiencies in ML processes to the technical details of building a Feature Store and how to integrate GCP with Snowflake using terraform. We demonstrate using the example of the Feature Store what risks are involved in not optimizing ML processes. This provides an easy-to-understand explanation of what MLOps is on the one hand, and on the other hand provides the technical details needed to adopt the solution right away. 

That's why we've divided the ebook into two parts: the business perspective and the technical perspective, so everyone can easily find the content that interests them. 

What will you find inside this MLOps ebook?

  • What MLOps is and the MLOps Platform
  • What impact MLOps implementation can have on business
  • 5 Machine Learning issues that can cause the ineffective use of data (along with a solution)
  • A step-by-step guide to building a Feature Store
  • Comparison of the most popular Feature Stores
  • Insight into Machine Learning - MLOps correlations
  • Example of MLOps architecture and workflow
  • How to integrate GCP with Snowflake using terraform
  • Vertex.ai platform - how it works in practice

See what the experts are saying about the ebook

This e-book describes the real-life challenges of running Machine Learning on production systems from both business and technical perspectives. Moreover, it guides you through the practical solutions, based on our expert's experience.  Check it out especially if you're struggling with latency, data silos, data drift or data skew!

Michał Bryś, Senior ML Engineer and Technical Product Owner

A very interesting position where the author describes the MLOps world and takes a look at it from a Feature Store perspective. This eBook contains solid theoretical information on what MLOps is and how Feature Stores can address its problems. The second part describes the first steps in Data Engineering work that will lead to creating the first features. This eBook provides very detailed information on how to start with dbt, Snowflake and Vertex AI tools.

I recommend it for everyone taking their first steps in the Data Engineering world or anyone who wants to extend his/her knowledge about Feature Stores.

Piotr Pękala, Project Architects Lead

Ideas are worthless without execution”. This popular adage can be also applied perfectly to Data Science and Machine Learning results, where so many companies jump at the idea of using ML, but are struggling to execute it by delivering ML apps to production. That’s why the MLOps movement was shaped, to gather all the engineering practices in one place to deploy and maintain ML models in production efficiently and reliably. A lot has been developed, especially recently around Feature Stores - how data is prepared, stored and fed to ML models in training and production.

In this e-book, Getindata engineers introduce the reader to the MLOps concept and then focus on demonstrating how to build a Feature Store using managed cloud services and open-source technologies. Unlike many vendor e-books, these guys spare no technical details of the architecture and the solution. While they show the solution based on particular technologies like Vertex AI, Snowflake or dbt, the knowledge and design are presented abstractly with replaceable components.

If you are responsible for the data & AI strategy in your company, this source of information can help you accelerate and grasp the concepts of MLOps and Feature Stores without a vendor selling you anything.

If you’re an engineer, this very juicy tutorial will show you in detail how you can implement a Feature Store at your company. It will not only introduce you to the problem and show how to do a PoC - from configuring cloud services to running your first pipeline, but will also help you productionize the solution with infrastructure as a code, CICD and so on. All you need from start to finish in a concise writeup!

Krzysztof Zarzycki, GetInData CTO

big data
machine learning
MLOps
Feature Store
27 October 2022

Want more? Check our articles

blogobszar roboczy 1
Success Stories

How we built a Modern Data Platform in 4 months for Volt.io, a FinTech scale-up

Money transfers from one account to another within one second, wherever you are? Volt.io is building the world’s first global real-time payment…

Read more
getindata nifi flow cicd notext
Tutorial

NiFi Ingestion Blog Series. PART II - We have deployed, but at what cost… - CI/CD of NiFi flow

Apache NiFi, a big data processing engine with graphical WebUI, was created to give non-programmers the ability to swiftly and codelessly create data…

Read more
managingmultipledatasourceobszar roboczy 1 4
Tutorial

Feature store - managing multiple data sources with Feast

As the effort to productionize ML workflows is growing, feature stores are also growing in importance. Their job is to provide standardized and up-to…

Read more
getindata cover nifi lego notext
Tutorial

NiFi Ingestion Blog Series. PART I - Advantages and Pitfalls of Lego Driven Development

Apache NiFi, big data processing engine with graphical WebUI, was created to give non-programmers the ability to swiftly and codelessly create data…

Read more
trucaller getindata control incoming calls cloud journey
Success Stories

Truecaller - armed with data analytics to control incoming calls

Building a modern analytics environment is a strategic, long-term, iterative process of continuous improvement rather than a one-off project. The…

Read more
transfer legacy pipeline modern gitlab cicd kubernetes kaniko
Tutorial

How we helped our client to transfer legacy pipeline to modern one using GitLab's CI/CD - Part 2

Please dive in the second part of a blog series based on a project delivered for one of our clients. If you miss the first part, please check it here…

Read more

Contact us

Interested in our solutions?
Contact us!

Together, we will select the best Big Data solutions for your organization and build a project that will have a real impact on your organization.

The administrator of your personal data is GetInData Sp. z o.o. Sp.k with its registered seat in Warsaw (02-508), 39/20 Pulawska St. Your data is processed for the purpose of provision of electronic services in accordance with the  Terms & Conditions. For more information on personal data processing and your rights please see  Privacy Policy.

By submitting this form, you agree to our Terms & Conditions and Privacy Policy