loading

Business Impact

  • Improved Operational Efficiency

  • Smart Insights and Visualization on the Performance of Models

  • Automated Production Model lifecycle management

  • Reducing the manual efforts for setup and configuration

Customer Key Facts

  • Country : India
  • Industry : Retail

Problem Context

The client is an Indian consumer goods company that primarily manufactures fashion accessories such as watches, jewelry, and eyewear.

They built Machine Learning models on local systems, which must be manually deployed and configured. The entire procedure took a long time, and the system was unscalable.

Challenges

  • Automating using Airflow by leveraging parameter store for around 100 models in production.
  • Building a generic template to accommodate over 100 models of different frameworks.
  • Rigorous testing of each model’s end-to-end pipeline in the development environment before deploying in the production environment.
Challenges

Technologies Used

Amazon EC2
AWS Lambda
Amazon ECR
Amazon SageMaker
Amazon Systems Manager
Amazon S3
Amazon CodeCommit
Amazon RDS
Amazon API Gateway
Amazon Athena
Amazon QuickSight
AWS Glue
Apache Airflow

Solution

Quantiphi helped the client in the orchestration of training, evaluation, and deployment of over 100 models running on SageMaker by leveraging Apache Airflow.

Once each model is deployed, the Production Model Lifecycle is such that the scoring, evaluation, and retraining of the model is automated.

Quantiphi also helped the client in the visualization of the performance of the deployed models using Amazon QuickSight and Amazon Athena.

Result

  • Reducing the manual efforts for setup and configuration.
  • Generalized templates for deployments with ease.

Looking for similar project?

Let's Talk

Get your digital transformation started

Let's Talk