Amazon SageMaker Unified Studio adds interactive interface for managing Feature Store in IAM Domains
Manage ML features visually in SageMaker Unified Studio — no API calls needed, with instant cross-tool sync via IAM role.
View original announcement →Visual Summary
graph TD
A{{Feature Store Interactive UI}}:::announced
B(SageMaker Unified Studio):::compute
C(SageMaker Feature Store):::storage
D(Amazon S3):::storage
E([Feature Discovery]):::feature
F([No-Code Management]):::feature
G([Ingestion Monitoring]):::feature
H((Data Scientists)):::external
I((Business Analysts)):::external
H ==>|"access"| B
I ==>|"access"| B
B ==>|"hosts"| A
A -->|"manages"| C
C -->|"offline store"| D
A -->|"enables"| E
A -->|"provides"| F
A -.->|"tracks"| G
C -.->|"IAM role sync"| A
classDef announced fill:#ff9900,stroke:#ec7211,color:#fff,font-weight:bold
classDef compute fill:#e3f2fd,stroke:#1565c0,color:#1565c0
classDef storage fill:#e8f5e9,stroke:#2e7d32,color:#2e7d32
classDef feature fill:#fff3e0,stroke:#e65100,color:#e65100
classDef external fill:#f5f5f5,stroke:#616161,color:#616161
What's New
Amazon SageMaker Unified Studio now includes a no-code, interactive interface for creating and managing feature groups in SageMaker Feature Store, available in IAM-configured domains. Data scientists, ML engineers, and business analysts can discover, create, modify, and monitor features entirely through a point-and-click UI without writing API calls or SDK code. Features created outside the Studio environment are automatically visible within it when the same IAM role is shared, enabling seamless cross-tool workflows.
How It Works
- Navigation: Users access Feature Store via the left navigation pane under AI/ML within their SageMaker Unified Studio project, landing on a page with two tabs — "Feature groups" and "Features."
- Feature Group Management: Users can create, view, edit, and delete feature groups directly from the project UI; each feature group is conceptually a table where columns are features and rows are records identified by a unique key.
- Storage Configuration: When creating a feature group, users choose between Online (low-latency, latest values only), Offline (full historical record in S3 as Parquet), or Online and Offline combined storage modes.
- Feature Discovery: The "Features" tab enables cross-group search, allowing users to find individual features across all feature groups without navigating into each group separately.
- Schema and Definition Viewing: Users can inspect feature definitions, data types, and schemas through the UI without querying the API.
- Ingestion Monitoring: Data ingestion status is visible in the interface; sample notebooks are provided for programmatic ingestion when needed.
- IAM Role Synchronization: Features created via SDK, notebooks, or other tools automatically appear in the Studio UI as long as the same IAM role is in use, ensuring consistency across environments.
Why It's Important
- Democratizes Feature Engineering: Business analysts and less code-fluent team members can now participate in feature management without needing Python SDK expertise, broadening ML team collaboration.
- Reduces Operational Friction: Eliminating boilerplate API calls for routine tasks like creating feature groups or checking ingestion status accelerates the ML development lifecycle.
- Single Pane of Glass: Consolidating feature management into SageMaker Unified Studio reduces context-switching between the console, notebooks, and CLI tools.
- Improved Feature Reuse: The cross-group feature search capability makes it easier to discover existing features before creating duplicates, promoting reuse and consistency across models.
- Faster Onboarding: New team members can explore and understand the feature store schema visually, shortening the time to productivity on ML projects.
How It's Different
- No-Code vs. SDK-Only: Previously, managing Feature Store required writing Python SDK or API calls; this UI eliminates that requirement for common operations, which was not available in the standard SageMaker console experience within Unified Studio.
- Unified Collaboration Environment: Unlike standalone Feature Store management in the SageMaker AI console, this interface is embedded directly in the project-centric Unified Studio workspace where data, notebooks, and models coexist.
- Cross-Tool Visibility: Features created via SDK or notebooks automatically sync into the UI via IAM role sharing — a bidirectional workflow not previously surfaced in a visual interface.
- IAM Domain Specificity: The interactive UI is currently scoped to IAM-configured domains, distinguishing it from SDK-based Feature Store access which works across all domain types.
- Integrated Feature Search: The dedicated "Features" tab for cross-group search is a new capability that goes beyond simple feature group listing, enabling feature-level discovery at scale.
When to Prefer It
- Cross-Functional Teams: When your ML team includes non-engineers or business analysts who need visibility into or control over feature definitions without writing code.
- Feature Auditing and Governance: When you need to quickly review feature schemas, definitions, and ingestion status across multiple feature groups for compliance or documentation purposes.
- Rapid Prototyping: When a data scientist wants to stand up a new feature group quickly during exploration without context-switching to a notebook just to write boilerplate SDK code.
- Feature Reuse Workflows: When onboarding a new project and you want to search existing features across all groups before deciding whether to create new ones.
- Collaborative ML Projects: When multiple team members share an IAM role and need a consistent, real-time view of the feature store regardless of how features were originally created.
- Monitoring Ingestion Pipelines: When you want a quick visual check on data ingestion status without querying CloudWatch or writing diagnostic code.
Availability
- GA Status: Generally available as of May 22, 2026.
- Domain Requirement: The interactive UI is available exclusively in Amazon SageMaker Unified Studio domains configured with AWS IAM; SDK-based Feature Store access remains available in all domain types.
- Regional Availability: Specific supported regions are not listed in the announcement; refer to the AWS Regional Services page and the SageMaker Unified Studio User Guide for current region support.
- Pricing: No separate pricing is announced for the UI itself; standard SageMaker Feature Store pricing applies for storage and reads/writes (online and offline store usage).
- Limitations: TTL configuration for online store record expiration is not supported via the UI and must be done through the SDK from a notebook; some advanced Feature Store capabilities may still require SDK usage as documented in the SageMaker AI Developer Guide.
