AWS and ThoughtWorks: How to get MLOps right

Taking machine learning from the labs into production, successfully.

Artificial intelligence and machine learning (ML) applications are becoming increasingly popular, however the process for developing, deploying, and continuously improving them is complex. As per VentureBeat report, 87% of projects never make it into production. MLOps extends DevOps into the machine learning space. MLOps refers to the culture where people, regardless of their title or background, work together to imagine, develop, deploy, operate, and improve a machine learning system.
In order to tackle the described challenges in bringing ML to production, ThoughtWorks has developed continuous delivery for machine learning (CD4ML), an approach to realize MLOps. In 2016, ThoughtWorks built a price recommendation engine with CD4ML on AWS for AutoScout24, the largest online car marketplace in Europe. Today, CD4ML is standard at ThoughtWorks for ML projects, because it has demonstrated success, and is one that incorporates modern, cloud-friendly software development practices. Together with AWS, we have created an ebook to guide you through an approach to get MLOps right.
Ebook: How to get MLOps right by ThoughtWorks and AWS

Download ebook

In most organizations, it takes between four months to a year to launch their first ML minimum viable product (MVP), according to the Harvard Business Review. Want to know how to tackle the complexity of building and deploying machine learning in your organization? Read this ebook to find out how.

Interested in our Data Strategy, Engineering and Analytics practice?