in Case studies
Upgrading AI into the production-ready stage with microservice-based MLOps
Case study details
A multinational company providing a complete marketing automation SaaS platform was already familiar with Machine Learning and Artificial intelligence but was looking for a way to bring its AI algorithm to the next production-ready level.
Addepto turned out to be the best partner to do that.
Challenge
Company AI models were implemented in two main product lines.
The first was especially crucial for the platform’s core feature, marketing automation.
With AI, the company could effectively verify e-mails and detect the ones that can be considered spam or phishing by webmails, so landing in SPAM folders never opened by end-users. A variety of AI models were designed to prevent that from happening, providing visibility on the desired level.
The second product – a website builder – included AI that enables end-users to migrate their existing web pages to the company’s solutions.
It made the whole process, which can be painful for non-technical users – smooth and easy: users could just put the link to their website, and AI – under the hood – do the magic.
The problem was that Machine Learning processes for both products were made from scratch as a single module.
With no MLOps “back office,” which slowed down the AI-driven processes. There were no pipelines that could automate, for example, the data flow, and the need for manually starting the operations significantly lowered their effectiveness.
The problem was that Machine Learning processes for both products were made from scratch as a single module.
With no MLOps “back office,” which slowed down the AI-driven processes. There were no pipelines that could automate, for example, the data flow, and the need for manually starting the operations significantly lowered their effectiveness.
Approach
Originally, just adding MLOps platform was planned, but refactorizing the AI models became necessary during the process
Addepto Team decided to revamp some of the AI models in order to avoid bottlenecks that might jeopardize the whole process in the following steps.
During project development, our team worked on:
- Addepto Team decided to revamp some of the AI models in order to avoid bottlenecks that might jeopardize the whole process in the following steps.
- This decision allowed for building a solid foundation of the MLOps platform based on modern microservice architecture, where every single element of the process can be replaced with no interference with the whole.
Goal
Addepto Delivered
The goal was to:
Rewrite the operating system to a separate CLI application
Build a retraining process
Develop MVP for a platform that actively monitors the quality of the standard MLOps model
Publish the model to the production environment
Outcome
Outcome
With MLOps, the company can actively monitor and optimize Machine Learning models in production.
By making adjustments as needed, managers can always be up-to-date with mail providers’ requirements, making the system efficient in delivering mail to end-users.
For example, the number of emails sent to a particular hosting provider never exceeds the allowable number, which prevents them from being considered SPAM. It helped the company to manage the risks and enabled them to scale up their capabilities to meet the needs of growing demand without sacrificing performance or reliability.
About Addepto
We assist businesses in leveraging AI technologies, including machine and deep learning, to improve their data utilization.
Here you can learn more about the technologies used in this project:
We enable the creation of data-driven services that enhance customer satisfaction, boost revenue, and expedite internal processes.
Let's discuss
a solution
for you
Edwin Lisowski
will help you estimate
your project.
- hi@addepto.com
Other case studies
Testing life cycles in manufacturing aerospace
Woodward is an independent designer, manufacturer, and service provider of energy control and optimization solutions for aerospace and industrial markets. The company was founded over 150 years ago. It was focused on delivering proven syste...
Compliance analysis at Retail
Spencer’s is a North American mall retailer with over 600 stores in the United States and Canada. Their stores offer gag gifts, clothing, band merchandise, sex toys, room decor, collectible figures, fashion and body jewelry, and fanta...