AIOps Modules
The Artificial Intelligence Operations (AIOps) modules provide specialized infrastructure components for building AI/ML workloads and data science platforms on AWS. These modules are designed to accelerate the deployment of machine learning infrastructure, model training, and AI operations.
Repository¶
Source: AWS AIOps Modules
Available Modules¶
Coming Soon - Feel free to directly access the Git Repository listed above in the meantime.
We will provide a comprehensive list of:
- current modules available
- description of each module
- input and output definitions
- example manifest configurations
Key Features¶
- ML-Focused: Purpose-built for machine learning and AI workloads
- Scalable: Designed for enterprise-scale ML operations
- Integrated: Works seamlessly with AWS AI/ML services
- Best Practices: Implements MLOps and AIOps best practices
- Production Ready: Battle-tested components for production ML systems
Usage Example¶
# Deploy a SageMaker Studio environment
name: ml-studio
path: git::https://github.com/awslabs/aiops-modules.git//modules/sagemaker/sagemaker-studio?ref=main&depth=1
targetAccount: primary
parameters:
- name: studio-domain-name
value: my-ml-platform
- name: enable-projects
value: true
Common Use Cases¶
- ML Platform Setup: Deploy complete SageMaker environments with notebooks, studios, and endpoints
- Model Training: Set up distributed training with Ray clusters on EKS
- MLOps Pipelines: Implement CI/CD for machine learning models
- Experiment Tracking: Deploy MLflow for experiment management and model registry
- Foundation Models: Deploy and fine-tune foundation models with Bedrock and SageMaker
- AI Agents: Build intelligent automation and autonomous systems
Getting Started¶
- Identify your ML use case from the categories above
- Review module documentation for specific AI/ML components
- Configure parameters for your ML workload requirements
- Deploy using Seed-Farmer for consistent, repeatable ML infrastructure
For more information about building ML platforms with Seed-Farmer, see the Module Development Guide.