Customer
Tier-2 universal bank with 45+ Bln $ AUM.
Challenge
Implement new bank-wide platform for ML models lifecycle automation.
Project Results
The new Data Science platform enabled quick implementation of ML-models of any complexity into business processes and eliminated operational risk associated with deployment of models into production environments. The deployed architecture of the platform automates the management of models, experiment tracking, code and data vesioning as well as model verification and deployment.
Business Value
Dramatic reduction of time-to-market for new ML-models, predominantly associated with credit scoring and risk monitoring. Improved performance of model-driven services thanks to the new state-of-the-art deployment architecture running on Kubernetes cluster.
Simplified onboarding of new data science team members, which now takes no more than 1 day.
Tech Stack
MLFlow, Gitlab, Minio, Jupyter, Kubernetes, Airflow.
Tier-2 universal bank with 45+ Bln $ AUM.
Challenge
Implement new bank-wide platform for ML models lifecycle automation.
Project Results
The new Data Science platform enabled quick implementation of ML-models of any complexity into business processes and eliminated operational risk associated with deployment of models into production environments. The deployed architecture of the platform automates the management of models, experiment tracking, code and data vesioning as well as model verification and deployment.
Business Value
Dramatic reduction of time-to-market for new ML-models, predominantly associated with credit scoring and risk monitoring. Improved performance of model-driven services thanks to the new state-of-the-art deployment architecture running on Kubernetes cluster.
Simplified onboarding of new data science team members, which now takes no more than 1 day.
Tech Stack
MLFlow, Gitlab, Minio, Jupyter, Kubernetes, Airflow.