Challenge

The project initiated with the setup of infrastructure using Terraform, defining and provisioning resources like VMs, networking, and storage on a chosen cloud platform. This phase ensured the scalable and efficient setup of resources required for model training and deployment.

Puppet was utilized for configuration management, ensuring all VMs were consistently configured with the necessary software and security settings. This phase ensured that the systems were secure and had the necessary dependencies for model training and deployment.

The model training and deployment phase involved creating a system to ingest and preprocess data, train AI models, and deploy them as API endpoints. This phase ensured that the AI models could be trained and updated efficiently and served to end-users with minimal latency.

Monitoring and logging were implemented using the ELK Stack, ensuring system health and activities could be monitored in real-time. This phase ensured that any issues could be identified and resolved promptly, and system usage could be analyzed for further optimization.