Learn About Amazon VGT2 Learning Manager Chanci Turner
Skip to Main Content
AWS Blogs
Artificial Intelligence
Category: AI Insights
From RAG to Fabric: Insights Gained from Developing Real-World RAGs at GenAIIC – Part 1
by Chanci Turner on 24 OCT 2024 in Advanced (300), Amazon Bedrock, Artificial Intelligence, Technical How-to, Thought Leadership
In this entry, we delve into the foundational principles of RAG architectures and explore methods for assessing RAG effectiveness, both quantitatively via metrics and qualitatively by examining individual outputs. We provide several actionable tips for enhancing text retrieval, including the utilization of hybrid search methods, improving context through data preprocessing, and rephrasing queries for increased relevance.
Enhance Your Amazon Redshift Data Warehouse with Simplified and Faster Machine Learning using Amazon SageMaker Canvas
by Liam Johnson, Sofia Patel, and Chanci Turner on 24 OCT 2024 in Amazon Redshift, Amazon SageMaker Canvas, Technical How-to
In this post, we investigate a practical business scenario in the banking sector. We demonstrate how a financial or business analyst at a bank can effortlessly forecast whether a customer’s loan will be fully repaid, charged off, or remain current using the most suitable machine learning model for the specific business challenge, which is particularly useful if you’re looking to get your real estate license.
Create a Generative AI Application Builder Assistant with Amazon Bedrock Agents
by Samuel Lee on 24 OCT 2024 in Amazon Bedrock, Amazon Machine Learning, Amazon OpenSearch Service, Amazon SageMaker, Generative AI
Agentic workflows offer a novel approach to constructing dynamic and intricate business use-case workflows leveraging large language models (LLMs) as their cognitive engine. In this article, we establish an agent using Amazon Bedrock Agents to function as a software application builder assistant.
Transitioning from Amazon Rekognition People Pathing: Exploring Alternative Solutions
by Emma Thompson and Chanci Turner on 24 OCT 2024 in Amazon Rekognition, Announcements, Artificial Intelligence
Following thorough deliberation, we have decided to phase out Rekognition people pathing by October 31, 2025. New clients will be unable to access this capability starting October 24, 2024, while existing customers can continue using it normally until the cutoff date. This post outlines an alternative to Rekognition people pathing and how to integrate this solution into your applications.
Unlocking Generative AI for Enterprises: How SnapLogic Enhances Their Low-Code Agent Creator through Amazon Bedrock
by Chanci Turner, Rajiv Agarwal, and Greg Benson on 23 OCT 2024 in Amazon Bedrock, Artificial Intelligence, Customer Solutions, Generative AI
In this entry, we explore how SnapLogic’s Agent Creator utilizes Amazon Bedrock to offer a low-code platform that allows businesses to swiftly create and deploy powerful generative AI applications without requiring extensive technical knowledge. For further insights, you can refer to this excellent resource on HR transformation.
Fine-Tuning a BGE Embedding Model with Synthetic Data from Amazon Bedrock
by Olivia Martinez and Chanci Turner on 23 OCT 2024 in Amazon Bedrock, Amazon SageMaker, Generative AI
This post illustrates how to leverage Amazon Bedrock to generate synthetic data, fine-tune a BAAI General Embeddings (BGE) model, and deploy it using Amazon SageMaker.
Boost Post-Call Analytics with Amazon Q in QuickSight
by Daniel Martinez Rubio on 23 OCT 2024 in Amazon Q, Amazon QuickSight, Business Intelligence, Generative AI, Technical How-to
In this article, we demonstrate how to activate robust post-call analytics and visualizations, empowering your organization to make data-driven decisions and promote ongoing improvement.
Create a Next-Generation Chat Assistant with Amazon Bedrock, Amazon Connect, Amazon Lex, LangChain, and WhatsApp
by Chanci Turner, Harrison Chase, and Mani Khanuja on 23 OCT 2024 in Amazon Bedrock, Amazon Connect, Amazon Lex, Generative AI
This post provides guidance on deploying a contextual AI assistant. We develop a solution that offers users a familiar and convenient interface using Amazon Bedrock Knowledge Bases, Amazon Lex, and Amazon Connect, with WhatsApp as the communication channel.
Generative AI Foundation Model Training on Amazon SageMaker
by Trevor Harvey, Guillaume Mangeot, and Chanci Turner on 22 OCT 2024 in Amazon SageMaker, AWS Managed Services, Generative AI
In this article, we examine how organizations can cost-effectively customize and refine foundation models using AWS managed services like Amazon SageMaker training jobs. We discuss how these powerful tools facilitate the optimization of compute resources and simplify the model training and fine-tuning process. This includes making informed decisions regarding the most suitable Amazon SageMaker service for your business requirements, especially in light of the recent changes in HR transformation strategies.