The New Amazon Sagemaker Walkthrough | Showing All Features

Why It Matters

The unified, no‑install SageMaker experience streamlines end‑to‑end AI development, letting enterprises prototype faster, reduce infrastructure overhead, and bring models to production with minimal friction.

Key Takeaways

• SageMaker Unified Studio launches with zero‑install, browser‑based interface.
• Integrated data catalog and S3 browsing simplify data source connections.
• AI agent generates Python or SQL code from natural language prompts.
• Visual ETL, workflows, and notebook environments support end‑to‑end pipelines.
• Built‑in model hub and inference endpoints enable rapid deployment.

Summary

The video walks viewers through Amazon SageMaker’s newly refreshed Unified Studio, a fully managed, browser‑based environment that requires no local installation. By logging in with existing IAM permissions, users instantly access data catalogs, S3 buckets, and a suite of connection options—including Glue, Redshift, Snowflake, and more—directly from the left‑hand navigation pane. Key features highlighted include on‑demand notebook instances where developers can install packages, switch compute types, and control costs via idle time‑outs. A standout AI agent translates natural‑language requests into executable Python or SQL code, exemplified by loading a CSV, displaying rows, and calculating top‑customer invoices. The interface also blends Python and SQL within a single notebook, enabling seamless data merges. The walkthrough demonstrates visual ETL builders, Airflow‑style workflow orchestration, and a query editor powered by Athena for ad‑hoc analysis. Additionally, SageMaker now bundles a model hub with popular LLMs (e.g., Mistral, Meta) and one‑click inference endpoint creation, accessible from notebooks or JupyterLab, and even integrates with VS Code for familiar development workflows. These enhancements dramatically lower the barrier to entry for data scientists and engineers, accelerating prototype to production cycles while consolidating data ingestion, transformation, model training, and deployment within a single, cost‑controlled AWS console.

Original Description

In this video, I do a quick walkthrough of the new Amazon SageMaker and show how everything works in the new unified interface.

We look at how SageMaker brings data analytics, notebooks, AI, and ML together in one place, with no setup required. I walk through accessing data from S3 and the AWS Glue Data Catalog, working with notebooks using both Python and SQL, using the built-in AI agent to generate code, and exploring data with charts.

You’ll also see visual ETL pipelines, query editing with Athena, model selection and deployment, inference endpoints, and how to work directly in JupyterLab or VS Code — all inside SageMaker.

If you’re curious how Amazon SageMaker looks and feels today, this is a fast end-to-end overview.

#dataengineering #aws #awspartner #sagemaker #ai #etl