Amazon Bedrock in SageMaker Unified Studio
Accelerate generative AI application development
Overview
Access Amazon Bedrock's capabilities through SageMaker Unified Studio to quickly build and customize your generative AI applications. This intuitive interface lets you work with high-performing foundation models (FMs) and use advanced features like Amazon Bedrock Knowledge Bases, Amazon Bedrock Guardrails, Amazon Bedrock Agents, and Amazon Bedrock Flows. You can develop generative AI applications faster within SageMaker Unified Studio's secure environment, ensuring alignment with your requirements and responsible AI guidelines.
Enable effortless generative AI development across skill levels
With Amazon Bedrock in SageMaker Unified Studio, you can develop generative AI applications through a simple, accessible experience designed for developers of all skill levels. The intuitive interface enables teams to collaborate effectively while using Amazon Bedrock's capabilities. You can access governed data, quickly prototype, iterate, and deploy production-ready generative AI apps based on your business needs.
Build custom generative AI applications
Customize FMs to match your requirements, data, workflows, and responsible AI standards. Create Knowledge Bases from your proprietary data sources using Retrieval Augmented Generation (RAG) to tailor your model responses to business needs. Build chat agent apps using Agents, add Guardrails for safeguarding and privacy, and leverage advanced features like prompt engineering, functions, and Flows – all without managing underlying services.
Collaborate seamlessly among stakeholders
Using Amazon Bedrock in SageMaker Unified Studio, you can collaborate seamlessly across business and technical teams, regardless of skill level. You can build, customize, and share generative AI applications securely, enabling trusted teamwork across functions. This helps your teams create various solutions, from company-specific content generation to workflow automation and software development.
Evaluate and adopt high-performing models with ease
Access a wide range of high-performing FMs from leading AI companies through the generative AI playground. You can compare different models and configurations to evaluate their performance easily. Using automated model evaluation, you can identify and select the best model for your use case based on performance, quality, and safety metrics.
Implement responsible AI guardrails
Create guardrails and set content filters on both user input and model responses to ensure appropriate outputs from your generative AI app. Customize guardrail behavior by adjusting filtering levels across categories and adding denied topics, aligning with your responsible AI guidelines and desired outputs.
Customers
-
Adastra
We build complex data analytics, ML and GenAI applications with built-in data governance and user-friendly interfaces. Before Amazon SageMaker Unified Studio, deploying multiple tools for our customers' data and information workers was mostly manual and time-consuming, and ensuring a robust data architecture provisioning was a challenge. Now, with Amazon SageMaker Unified Studio, we can deploy a single data worker tool for data engineers and ML scientists. We are also automating data infrastructure deployment, allowing us to simplify the process for our customers and enhance their experience.
Zeeshan Saeed, Chief Technology and Strategy Officer, Adastra -
Toyota Motor North America
To address siloed data sets spread across our automotive operations, we are implementing Amazon SageMaker to help unify and govern data across our connected car, sales, manufacturing, and supply chain units. This approach allows us to search, discover, and share data effortlessly, laying the groundwork to pre-empt quality issues, increase customer satisfaction, and enable easier development of generative AI applications.
Kamal Distell, VP of Data, Analytics, Platforms, and Data Science, TMNA