Case Study: Standardizing FMCG Operations with the MLOps best practices Across Multiple Cloud Platforms

Our client faced the challenge of expanding various ML use cases across multiple cloud platforms. They sought a solution that would standardize their operations, streamline ML training, and inference tasks while ensuring flexibility and scalability.

Our team embarked on a transformative project to standardize the client’s operations using MLOps templates and best practices for training and inference across GCP, Azure, and Databricks. Leveraging the power of cloud technologies and AI, we designed and implemented a cloud-agnostic tool and a customized MLOps template that were seamlessly integrated in ML use cases for different cloud platforms, while ensuring maintenance and accelerated development.

A cloud-agnostic tool and MLOps template that allows to create, trigger, and schedule ML training and inference pipelines across multiple cloud platforms.

Key implementations:

• Automated process of CI/CD that meets organization’s standards and best practices.
• Resource autoscaling and consumptions logging.
• Model versioning, model monitoring, and model retraining pipelines (triggered if data / concept drift is detected).
• Deployment of a Q&A chat using ChatGPT and Langchain to provide support and guidance for data scientists navigating through the cloud-agnostic tool and MLOps template.

The standardization of various use cases using the cloud-agnostic tool and MLOps template resulted in tangible benefits for our client. Maintenance tasks were significantly streamlined, leading to increased operational efficiency and reduced overhead costs. By implementing CI/CD pipelines with Github Actions, we accelerated the development cycle and seamlessly transitioned use cases from development to production. Furthermore, the deployment of a chatbot provided users with real-time support and guidance, enhancing user satisfaction and productivity

Contact us to see how we can help you.
We’ll get back to you within 4 hours on working days (Mon – Fri, 9am – 5pm).