Skip to content
For Self-Managed AI Platform users running v11.1, see the on-premise platform documentation

Installation and configuration guide

The DataRobot platform is an enterprise-grade AI solution that unifies the end-to-end machine learning lifecycle for both predictive and generative AI workloads. Designed for data scientists, ML engineers, and IT professionals, it provides both an intuitive, low-code GUI and robust, programmatic access via REST APIs and integrated notebooks.

The platform leverages powerful AutoML to automate complex tasks, accelerating data preparation, model building, and hyperparameter tuning by testing hundreds of model blueprints in parallel. Architecturally, DataRobot is cloud-native and highly scalable, deployed on Kubernetes clusters using Docker containers. Key modules include the application server, dedicated modeling workers, and prediction servers. It provides robust MLOps capabilities for governance, monitoring, and auditability, ensuring seamless integration with S3-compatible storage, enterprise data warehouses, and external tools like MLflow and Airflow.

This guide outlines the end-to-end process for installing, configuring, and maintaining your self-managed DataRobot installation. The process is divided into several phases. Follow each phase in order, using the links to access detailed instructions for each step.

Phase 1: Pre-installation planning

Before you begin any hands-on setup, it is critical to review all requirements and plan your environment. This phase answers the question, "What do I need to know and have ready before I start?"

Special case requirements and considerations:

DataRobot relies on several Persistent Critical Services (PCS), including PostgreSQL, Redis, and MongoDB. You can deploy these services within the Kubernetes cluster or use external, cloud-managed services.

For more information on each option, see the following guides:

Phase 2: Installation

Platform-specific installation and configuration guides:

Phase 3: Post-installation and verification

After the main installation commands are complete, follow these guides to verify that the platform is operating correctly. This phase answers the question, "How do I know if it worked?"

Phase 4: Advanced configuration

This phase covers optional, advanced configurations that allow you to fine-tune your DataRobot installation.

備考

Many of these advanced settings are configured by adding or modifying values in your values_dr.yaml file. As such, you should review these topics during Phase 2 while preparing your Helm chart values.