#754: Accelerating healthcare decisions with agents

In this episode of the AWS Podcast, you'll hear how Cohere Health®, uses Amazon Bedrock AgentCore to optimize the accuracy and efficiency of health plan medical necessity reviews. Cohere Review Resolve™ analyzes both structured and unstructured data–such as clinical records, patient notes, and faxes, quickly identifying and surfacing evidence to validate the medical necessity of a requested treatment. Hear why Cohere Health chose Amazon Bedrock AgentCore for their production deployment of agentic AI in a highly regulated healthcare environment and how this integration in Review Resolve™ reduces review times helping them to meet critical, mandated turnaround times.

In this episode, Jillian speaks with Cohere Health®, a clinical intelligence company focused on strengthening payer-provider collaboration as well as improving the speed and accuracy of clinical decision-making, both pre- and post-care. The company’s clinically trained AI helps accelerate access to patient care, improve patient outcomes, reduce administrative burden for providers, and improve healthcare economics across the care continuum. Using Amazon Bedrock AgentCore, Cohere Health built Cohere Review Resolve™, an AI-powered copilot that optimizes the accuracy and efficiency of health plan medical necessity reviews. Cohere Review Resolve™ analyzes both structured and unstructured data–such as clinical records, patient notes, and faxes, quickly identifying and surfacing evidence to validate the medical necessity of a requested treatment. This conversation reveals why Cohere Health chose Amazon Bedrock AgentCore for their first production deployment of agentic AI in a highly regulated healthcare environment. Cohere Health expects that Review Resolve™ will reduce review times by 30-40%, helping them to meet critical, mandated turnaround times. For patients, quicker decision-making will accelerate care access, increase adherence to therapy, improve outcomes and reduce costs. With Amazon Bedrock AgentCore, Review Resolve™ also improves upon their existing 90% automation rate while helping health plans by further increasing the accuracy of clinical determinations, thereby improving medical expense savings and patient outcomes.

In this episode, Simon speaks with Joe Magerramov (VP & Distinguished Engineer) to explore the transformative impact of AI-assisted coding on software development workflows. Joe shares his team's real-world experience achieving a 10x increase in code throughput using agentic development, but warns that simply bolting AI agents onto existing practices is like "adding a turbocharger to a car with narrow tires and old brakes." We dive deep into the critical infrastructure changes needed to sustain high-velocity development, including the mathematics of bug probability at scale, innovative testing approaches inspired by aviation industry practices, and the evolution of CI/CD pipelines that can handle dozens of commits per hour rather than per day. The conversation reveals why the biggest opportunity isn't just writing more code faster, but using AI to make previously impractical engineering practices economically viable—from comprehensive end-to-end testing with fake dependencies to rapid feedback loops that prevent the entire development pipeline from grinding to a halt when issues arise. https://blog.joemag.dev/2025/10/the-new-calculus-of-ai-based-coding.html

Discover how customer are leveraging AWS to modernize SAP systems. Tushar Srivastava (Principal Account Manager) talks about the role SAP plays in an enterprise, the modernization challenges faced by customers and how AWS helps in that modernization journey. Through the conversation we will learn about the options that exist and how can customers make smart choices and leverage all that AWS has to offer in the context of SAP systems. Tushar has written a book ‘Modernizing SAP with AWS’, published by Springer Nature, which takes the reader on an SAP Modernization journey through the lens of ‘Nimbus Airlines’, a fictional company running SAP. Join us for an engaging conversation around SAP Modernization, Tushar’s motivation in writing the book and some key learnings for customers looking to embark on their own SAP modernization program. https://a.co/d/9auwYxf

For the past 6 years, Werner has published his annual tech predictions, where he’s covered everything from sports and simulation, to smart energy innovation and quantum to AI-supported software development, even the role companies will play in educating the next generation of engineers. I sat with Werner to discuss why he makes predictions and what he sees coming in 2026 and beyond. You’re in for a treat. Read the full predictions on www.allthingsdistributed.com .

It is the end of re:Invent! Simon and Jillian share some updates and also take a moment to reflect on 2025.

Simon and Jillian catch you up on the highlights from today's keynote PLUS all the "pre:Invent" announcements that took place prior to the event!

In this episode, Matt Garman's 2025 re:Invent keynote unveils exciting AI advancements, including Amazon Nova to Lite, a cost-effective reasoning model, and Amazon Nova 2 Sonic, a new speech-to-text model. The keynote also covers Security, Storage, Compute, Networking, and a whole lot more!

Discover how AWS leverages automated reasoning to enhance AI safety, trustworthiness, and decision-making. Byron Cook (Vice President and Distinguished Scientist) explains the evolution of reasoning tools from limited, PhD-driven solutions to scalable, user-friendly systems embedded in everyday business operations. He highlights real-world examples such as mortgage approvals, security policies, and how formal logic and theorem proving are used to verify answers and reduce hallucinations in large language models. This episode delves into the exciting potential of neurosymbolic AI to bridge the gap between complex mathematical logic and practical, accessible AI solutions. Join us for a deep dive into how these innovations are shaping the next era of trustworthy AI, with insights into tackling intractable problems, verifying correctness, and translating complex proofs into natural language for broader use. https://aws.amazon.com/what-is/automated-reasoning/

There are so many updates this week you might need two cups of coffee! Simon and Jillian guide your way.