AWS re:Think Podcast

AI can still sometimes hallucinate and give less than optimal answers. To address this, we are joined by Arize AI's Co-Founder a Aparna Dhinakaran for a discussion on Observability and Evaluation for AI. We begin by discussing the challenges AI Observability and Evaluation. For example, how does “LLM as a Judge” work? We conclude with some valuable advice from Aparna for first time entrepreneurs.Begin Observing and Evaluating your AI Applications with Open Source Phoenix:https://phoenix.arize.com/AWS Hosts: Nolan Chen & Malini ChatterjeeEmail Your Feedback: rethinkpodcast@amazon.com

In this episode we are going to discuss Model Context Protocol (MCP) in context of Agentic architecture with Mona - Gen AI Specialist Solutions Architect at AWS.We will cover the challenges of implementing Agents and how MCP can help. We will try to simplify it with an example of MCP with Agents and further delve deep into LangGraph adapters such as FastMCP and FastAPI.You will also learn about implementing MCP architecture on AWS.Apart from her daily job, Mona has published two books Natural Language Processing with AWS AI Services: Derive strategic insights from unstructured data with Amazon Textract and Amazon Comprehend and Google Cloud Certified Professional Machine Learning Study Guide. She has authored multiple on AI/ML and cloud technology and a co-author on a research paper on CORD19 Neural Search. She is also a frequent speaker at multiple conferences such as ISMB 2022, AWS Re:Invent. Links https://www.amazon.com/Natural-Language-Processing-AWS-Services/dp/1801812535https://a.co/d/7rlkZwnBlog - Implement MCP in SageMaker AI https://aws.amazon.com/blogs/machine-learning/extend-large-language-models-powered-by-amazon-sagemaker-ai-using-model-context-protocol/Blog - Implement MCP with Bedrock Agents https://aws.amazon.com/blogs/machine-learning/harness-the-power-of-mcp-servers-with-amazon-bedrock-agents/AWS Hosts: Nolan Chen & Malini ChatterjeeEmail Your Feedback: rethinkpodcast@amazon.com

Join Jobi George and Tony Le from Weaviate as we discuss how their AI Native Vector Database is powering modern AI applications. We begin with an intro to Vector Databases followed by common use cases, and how they fit into Agentic Workflows. We also talk about the advantages of using an AI Native Database from the Open Source community. To learn more:Website - www.weaviate.ioLinkedin - https://www.linkedin.com/company/weaviate-io/AWS Marketplace - https://aws.amazon.com/marketplace/seller-profile?id=seller-jxgfug62rvpxsDeveloper Community - https://weaviate.io/developers/weaviate/model-providers/awsInstagram - https://www.instagram.com/weaviate.io/Youtube - https://www.youtube.com/@WeaviateAWS Hosts: Nolan Chen & Malini ChatterjeeEmail Your Feedback: rethinkpodcast@amazon.com

Join Julien Simon, Chief Evangelist of Arcee AI, as we discuss how Model Merging can help us build more versatile and cost effective AI models. Hear him explain the different methods for Model Merging while comparing it with other techniques for building better models such as Model Distillation and Transfer Learning. To learn more about Model Merging:Arcee AI Websitehttps://arcee.aiArcee's MergeKit: A Toolkit for Merging Large Language Modelshttps://arxiv.org/abs/2403.13257Deep dive: model merging on YouTube:https://youtu.be/cvOpX75Kz4M?si=nT2U7Fe8oLNRSy0uTry MergeKit:https://github.com/arcee-ai/mergekitAWS Hosts: Nolan Chen & Malini ChatterjeeEmail Your Feedback: rethinkpodcast@amazon.com

Your AI is only as good as the data its trained on. In this episode we talk to Krishna Subramaniam, Co-Founder and President of Komprise, a provider of intelligent Data Management. We discuss both the importance and challenge of unlocking the value unstructured data for AI pipelines. For more info on Komprise's Intelligent Data Management Platform:www.komprise.comAWS Hosts: Nolan Chen & Malini ChatterjeeEmail Your Feedback: rethinkpodcast@amazon.com

"2025 is the year of inference." Todd Pond, AWS Director of Strategic Sales, is out in the field with commercial customers every day, helping them leverage technology to solve their biggest problems, accelerate innovation and transform their businesses with AI. He and his team have a deep understanding of these companies - the challenges they're up against, their limited resources, their unique value propositions, and how to enable their growth with maximum ROI. In this episode, we are happy to have Todd back again to discuss what were the 2024 trends, what were the pleasant surprises, his actionable insights and some recipe for success in 2025 and beyond.

Join AWS Head of AI and Modern Data Strategy Business Development Ben Schreiner for a recap of re:Invent 2024. Ben share his top 5 themes from the conference as well as what surprised him in 2024 and what he thinks lies ahead in tech in 2025. AWS Hosts: Nolan Chen & Malini ChatterjeeTop Announcements from re:Invent 2024:https://aws.amazon.com/blogs/aws/top-announcements-of-aws-reinvent-2024/https://aws.amazon.com/blogs/machine-learning/aws-reinvent-2024-highlights-top-takeaways-from-swami-sivasubramanian-to-help-customers-manage-generative-ai-at-scale/Email Your Feedback: rethinkpodcast@amazon.com#awsrethink

In this episode, Tailscale CEO Avery Pennarun joins us to discuss VPNs, network connectivity and security in the cloud era. We talk about hub & spoke vs full mesh networks as well current and future trends in networking. Avery also talks about how Tailscale's approach to networking helps make setting up secure networks between devices and the cloud easier than before. For more info on Tailscale check out:tailscale.comAWS Hosts: Nolan Chen & Malini ChatterjeeEmail Your Feedback: rethinkpodcast@amazon.com

In this AWS re:Think Podcast episode, Rahul Bhargava reveals how Crop.photo by Evolphin is revolutionizing e-commerce visuals with AI magic. Discover how smart cropping and background generation are transforming product images from mundane to magnificent. Rahul discusses the technology behind Crop.photo's ability to enhance online stores, from fashion to furniture, creating eye-catching digital storefronts. Tune in for an exciting glimpse into the future of Gen AI-driven bulk image editing that's reshaping the online shopping experience.AWS Hosts: Vaishnavi Ganesan & Malini ChatterjeeEmail Your Feedback: rethinkpodcast@amazon.comGet Started for Free: https://app.crop.photo/user/signup

In the rapidly evolving world of cloud computing, financial operations (FinOps) has emerged as a critical discipline for organizations looking to optimize their cloud spending and mitigate financial risks. In this episode of the AWS re:Think podcast, we dive deep into the latest advancements in the FinOps space, exploring how organizations can navigate the financial complexities of cloud deployment and management. Joining us is Aran Khanna, the Co-founder and CEO of Archera.ai, a leading provider of FinOps solutions. Aran shares his insights on the importance of FinOps, the latest tools and strategies, and how Archera's innovative offerings, including their "Insured Commitments" program, can help customers mitigate the financial risks associated with the cloud. Whether you're in the midst of a cloud migration or looking to optimize your existing cloud infrastructure, this episode is a must-listen for anyone looking to rethink their cloud strategy and maximize the financial benefits of the cloud. Check out Archera.ai at https://archera.ai/ to learn more.AWS Hosts: Nolan Chen & Anu JayanthiEmail Your Feedback: rethinkpodcast@amazon.com

Companies now have a wide array of choices when it comes to Large Language Models. In this episode we talk to Ivan Lee, CEO of datasaur.ai about choosing the right LLM for your AI projects. We discuss open source vs proprietary models and the metrics for evaluating these models. We also cover Small Language Models as well and when to use more specialized vs general purpose models.Datasaur:https://datasaur.ai/LLM Labs from Datasaur:https://datasaur.ai/llm/homeAWS Hosts: Nolan Chen & Malini ChatterjeeEmail Your Feedback: rethinkpodcast@amazon.com

Companies need to provide real time insights to both customers and internal users. These insights power use cases such as personalization and fraud detection. StarTree Cloud is a real-time analytics platform built on Apache Pinot for building such applications that depend on real time insights. In this episode we meet with Chinmay Soman, Head of Product at Startree.ai to discuss the different dimensions of real-time analytics and how Apache Pinot and StarTree Cloud offer a robust platform for providing such insights to applications.AWS Hosts: Nolan Chen & Malini ChatterjeeEmail Your Feedback: rethinkpodcast@amazon.comResources:StarTree:https://startree.aiStarTree community Slack:https://communityinviter.com/apps/startreedata/startree-communityApache Pinot Slack: https://communityinviter.com/apps/apache-pinot/apache-pinotServerless / Free forever workspace:https://stree.ai/free

Today we join Maria Khalusova, Staff Developer Advocate with Unstructured.IO, to discuss how companies can unlock their Unstructured Data to deliver better results from their Large Language Models. We talk about how Unstructured Data can enhance the performance of RAG applications, RAG vs Fine Tuning, data Chunking, Multi-Modal models and more. AWS Hosts: Nolan Chen & Malini ChatterjeeUnstructured Enterprise Platform beta signup: https://unstructured.io/platformEmbedding models MTEB Leaderboard: https://huggingface.co/spaces/mteb/leaderboard2019 Deloitte report (source of the statistics that only 18% of organizations were using unstructured data):https://www2.deloitte.com/us/en/insights/topics/analytics/insight-driven-organization.html80% of data is unstructured, source: https://mitsloan.mit.edu/ideas-made-to-matter/tapping-power-unstructured-dataPapers showing RAG outperforming fine-tuning: https://arxiv.org/abs/2312.05934https://arxiv.org/abs/2401.08406Email Your Feedback: rethinkpodcast@amazon.com

In this podcast episode, we explore with Dr. Alok Aggarwal, CEO and Chief Data Scientist at Scry AI. about the past, present, and future of artificial intelligence. He traces the evolution of AI from its early beginnings in the 1950s to its current state, marked by rapid advancements and the emergence of the fourth industrial revolution. We listen to the story behind Scry AI. and how we explore the world of generative AI.If you are interested to learn more about his book "The Fourth Industrial Revolution & 100 Years of AI (1950-2050)”, here is the link to purchase it from Amazon: https://a.co/d/arodd1HAWS Hosts: Nolan Chen & Malini ChatterjeeEmail Your Feedback: rethinkpodcast@amazon.com

In this podcast episode, we delve into the crucial topic of security related to AI and Machine Learning, with a particular focus on Generative AI. As AI and ML technologies rapidly advance, it is imperative to implement a robust security strategy based on the principle of Defense in Depth. We'll explore the potential security and privacy risks associated with AI/ML systems, highlighting the importance of Responsible AI practices. Additionally, we'll discuss practical approaches to implementing security measures within specific AI/ML services, such as Amazon Bedrock, a secure and compliant foundation for building and deploying AI/ML applications on AWS. Join us as we navigate the intricate landscape of AI/ML security, equipping you with the knowledge and best practices to safeguard your AI/ML deployments and mitigate potential risks.AWS re:Inforce 2024:https://hub.reinforce.awsevents.com/attendee-portal/catalog/?filters=5B4D572F-79F0-EE11-81DE-9CF328ECC866&search=APS351AWS Hosts: Nolan Chen & Malini ChatterjeeEmail Your Feedback: rethinkpodcast@amazon.com

The amount of streaming data for Observability continues to grow exponentially from GBs to TBs daily. Join Edge Delta's CEO Ozan Unlu as we discuss how their AI Observability Platform can deliver more Sustainable Observability, and why the needs of different size data volumes from different industries requires us to rethink Observability architecture in the Cloud. Ozan also offers valuable career advice for fast changing times.AWS Hosts: Nolan Chen & Malini ChatterjeeEmail Your Feedback: rethinkpodcast@amazon.com

Many companies are sampling away data and collecting only trends. Join Axiom's Chief Extrovert Kendall Miller as he explains why sampling your data can disrespect your individual users. AWS Hosts: Nolan Chen & Malini ChatterjeeKendall Miller: https://www.linkedin.com/in/kendallamiller/Email Your Feedback: rethinkpodcast@amazon.com

In this episode we are joined by one of our AWS customers, Mixbook with DJ Charles, CTO and David Newhoff, CPO to share how they are leveraging generative AI to empower creativity and deepen human connection by unlocking the future of storytelling. Join us to hear how Mixbook took an idea from concept to production to empower their end-user's creativity with AI-driven recommendations for images, backgrounds, stickers, captions, and more.Try out the app: https://www.mixbook.com/AWS Hosts: Nolan Chen & Malini ChatterjeeEmail Your Feedback: rethinkpodcast@amazon.com

In this episode, we join AWS Solutions Architects, Krishna Adettiwar and Eric Kiesow, to discuss how leveraging a multi-account strategy can help you protect your workloads, increase pace of innovation, and give you more cost visibility.AWS Hosts: Nolan Chen & Malini ChatterjeeEmail Your Feedback: rethinkpodcast@amazon.comVisit these resources to get started with the multi-account strategy:Organizing Your AWS Environment Using Multiple Accounts (AWS Whitepaper) - https://docs.aws.amazon.com/whitepapers/latest/organizing-your-aws-environment/organizing-your-aws-environment.htmlMulti-account strategy for small and medium businesses (AWS blog) - https://aws.amazon.com/blogs/mt/multi-account-strategy-for-small-and-medium-businesses/AWS Cloud Operations & Migrations Blog - https://aws.amazon.com/blogs/mt/

In this episode we meet with Matthew McClean, a Sr Manager from AWS's Annapurna Team to talk about Accelerators - the chips that make AI possible. We cover different Accelerators - GPUs, Trainium, Inferentia, Graviton and more.AWS Hosts: Nolan Chen & Malini ChatterjeeEmail Your Feedback: rethinkpodcast@amazon.comLinks for the Show:AWS re:Invent 2023 - Behind-the-scenes look at generative AI infrastructure at Amazon (CMP206)https://www.youtube.com/watch?v=fDk09hms8s8AWS re:Invent 2023 - Behind-the-scenes look at generative AI infrastructure at Amazon (CMP206)https://www.youtube.com/watch?v=VPvguzeWlbUWelcome to AWS Neuronhttps://awsdocs-neuron.readthedocs-hosted.com/en/latest/index.htmlOptimum Neuronhttps://huggingface.co/docs/optimum-neuron/index

Many of the challenges we see that customers experience on their cloud journey are non-technical. Leveraging your AWS Account team and our vast AWS Partner Ecosystem can help, but AWS also offers programs like Experience-Based Acceleration (EBA). EBA is a transformation methodology using hands-on agile, immersive interactions that build cloud capability and accelerate business value from cloud adoption.AWS Hosts: Nolan Chen & Malini ChatterjeeEmail Your Feedback: rethinkpodcast@amazon.comLearn more about EBA:https://aws.amazon.com/blogs/mt/level-up-your-cloud-transformation-with-experience-based-acceleration-eba/

HPC is more than just high performance computing. It's also high performance storage and networking. In this episode we discuss how HPC is used to solve important problems in Computational Fluid Dynamics, Genomics, Medicine, AI and more. We also talk about how AWS services such as ParallelCluster, EFA and FSx for Lustre make it easier to deploy high performance workloads in the cloud. AWS Hosts: Nolan Chen & Malini ChatterjeeEmail Your Feedback: rethinkpodcast@amazon.comHPC on AWS Links:https://aws.amazon.com/hpc/https://day1hpc.com/#awsrethink

Join AWS Business Innovation leader Ben Schreiner for a recap of re:Invent 2023. We cover major announcements and innovations in infrastructure, data analytics and generative AI. AWS services discussed include S3 Express One Zone, Graviton 4, Bedrock, Amazon Q, Zero ETL and more!AWS Hosts: Nolan Chen & Malini ChatterjeeFurther reference - https://www.aboutamazon.com/news/aws/aws-reinvent-2023-announcementsTop announcements - https://aws.amazon.com/blogs/aws/top-announcements-of-aws-reinvent-2023/Email Your Feedback: rethinkpodcast@amazon.com#awsrethink

In this episode we join AWS Solutions Architect Vihang Shah to talk about strategies to optimize costs and performance with AWS Lambda. We will share insights into real world scenarios and provide actionable tips to strike a balance between performance efficiency and cost savings in the realm of Serverless computing. Its time to optimize your Serverless journey with AWS Lambda!AWS Hosts: Nolan Chen & Malini ChatterjeeEmail Your Feedback: rethinkpodcast@amazon.comLinks for the show: Learn about Serverless: https://aws.amazon.com/training/learn-about/serverless/Fundamentals of AWS Lambda: https://serverlessland.com/content/service/lambda/guides/aws-lambda-fundamentals/what-is-aws-lambdaServerless Optimization Workshop (Performance and Cost): https://catalog.us-east-1.prod.workshops.aws/workshops/2d960419-7d15-44e7-b540-fd3ebeb7ce2e/en-US

We join AWS Principal Networking Specialist and author James Devine to talk about building Wide Area Networks In the Cloud. We begin by covering the basics of WANs. We then talk about MPLS, VPNs and SD-WAN. We also cover AWS services for building WANs such as CloudWAN, Transit Gateway, SiteLink and more. AWS Hosts: Nolan Chen & Malini ChatterjeeEmail Your Feedback: rethinkpodcast@amazon.comLinks for the show:https://aws.amazon.com/blogs/networking-and-content-delivery/build-global-sd-wans-with-aws-cloud-wan-tunnel-less-connect/https://aws.amazon.com/cloud-wan/https://aws.amazon.com/vpn/https://aws.amazon.com/directconnect/https://aws.amazon.com/blogs/networking-and-content-delivery/introducing-aws-direct-connect-sitelink/https://www.youtube.com/watch?v=flBieylTwvI#awsrethink

"We are enabling a much wider adoption of playing-field-leveling technology." Todd Pond, AWS Director of Strategic Sales, is out in the field with commercial customers every day, helping them leverage technology to solve their biggest problems, accelerate innovation and transform their businesses. He and his team have a deep understanding of these companies - the challenges they're up against, their limited resources, their unique value propositions, and how to enable their growth with maximum ROI. In this episode, we talk to Todd about his career journey as a long-time sales leader and get his actionable insights on the biggest opportunities for success in 2024 and beyond.

In this episode we are joined by AWS Developer Advocate and longtime networking professional Brandon Carroll to discuss how computer networking has evolved over the years, not just for companies but for networking professionals as well. We talk about the challenges of deploying, monitoring and securing networks both on-prem and in the cloud. Brandon also shares thoughts on what it would take for young people to dive into computer networking today and in the future.AWS Hosts: Nolan Chen & Malini ChatterjeeEmail Your Feedback: rethinkpodcast@amazon.com

Join AWS Go To Market leader Tim Henderson as he introduces us to Amazon's Graviton processor. Learn how Graviton is able to deliver better performance with less cost. We also discuss the types of cloud workloads that are best suited for running on Graviton. We conclude with resources to help you get started with Graviton:https://aws.amazon.com/ec2/graviton/fast-start/https://github.com/aws/porting-advisor-for-gravitonAWS Hosts: Nolan Chen & Malini ChatterjeeEmail Your Feedback: rethinkpodcast@amazon.com

In this episode we join AWS Sr SA Lavanya Bandari to talk about how customers can get ready for Generative AI. We talk about common use cases for Gen AI, and how they can now be easily deployed with Amazon Bedrock. We also discuss when fine tuning of Large Language Models is needed. AWS Hosts: Nolan Chen & Malini ChatterjeeEmail Your Feedback: rethinkpodcast@amazon.com

What is Data Monetization? Why is it critical to think about data monetization in the context of business strategy? In this episode we join AWS Sr Solution Architects Priyanka Sadhu and Robert Daly to talk about the basics of data monetization. We discuss how it can be used with associated processes to generate measurable business value and cover some of the best practices to get started.AWS Hosts: Nolan Chen & Malini ChatterjeeEmail your feedback! rethinkpodcast@amazon.com

Join AWS Sr Solutions Architect Rohan Phadke for a conversation on optimizing storage costs in the cloud. When using cloud storage, it is important to take control of your costs and optimize your spend while building modern, scalable applications. In this podcast, learn about different strategies to optimize costs when using cloud based storage solutions. AWS Hosts: Nolan Chen & Malini Chatterjee

In this episode we join AWS SAs Krati Singh and Patrick Duffy to discuss approaches to securing your workloads in the Cloud. We cover the Shared Responsibility Model, Identity & Access Management and Infrastructure & Data Protection.AWS Hosts: Nolan Chen & Malini ChatterjeeEmail us your feedback! rethinkpodcast@amazon.com

Do you know AIOps? Would you like to better automate your IT Operations?Join AWS SAs Mohit Gadkari and Archana Ambavane as they talk about how AI/ML services can be leveraged to add additional intelligence to parse through all the noise and surface only the appropriate signals from your IT Operations.AWS Hosts: Nolan Chen & Malini ChatterjeeEmail us your feedback! rethinkpodcast@amazon.com

Join AWS SA Mahesh Biradar as he discusses Observability in the Cloud. Observability is the capability to continuously generate and discover actionable insights based on signals from the system under observation. In other words, it allows users to understand a system's state from its external output and take (corrective) action. The observability of a system has a significant impact on its operating and development costs. Observable systems yield meaningful, actionable data to their operators, allowing them to achieve favorable outcomes (faster incident response, increased developer productivity) and less toil and downtime.AWS Hosts: Nolan Chen & Malini Chatterjee

In this episode we talk about Containers in the cloud with AWS Solutions Architects Jamal Arif and Nikita Patel. We start with the basics - What Are Containers? When does it make sense to use Containers? We then dive into Docker, Docker Swarm, Kubernetes and how container orchestration can be implemented through specific AWS services such as EKS and Fargate. If you're new to Containers, this episode will help you decide the right container strategy in the cloud.AWS Hosts: Nolan Chen & Malini Chatterjee

What is “Serverless” Computing? In this episode we explore what Serverless Computing is with AWS SAs Sabha Parmaseswaran and Hardik Vasa. We discuss what companies should consider when transitioning from an existing architecture to this newer cloud native approach to computing. We also cover related AWS Services such as Lambda and API Gateway as well as Security and Costs.AWS Hosts:Nolan ChenMalini Chatterjee

Cost Optimization is top of mind in uncertain economic times. In this episode we join AWS SAs Krati Singh and Mike Tuszynski to discuss how Right Sizing your cloud resources with the right tools can help you ensure sufficient capacity without overspending.AWS Hosts: Nolan Chen & Malini ChatterjeeEmail us your feedback! rethinkpodcast@amazon.com

In this episode we begin by asking ‘What is Generative AI?' We meet with AWS Senior SA Michael Nguyen to talk about Foundation Models and Large Language Models. We then dive into Gen AI use cases and how businesses can use the cloud to build innovative Gen AI solutions.AWS Hosts:Nolan ChenMalini Chatterjee

What is a Solutions Architect? What do they do? And what are they NOT? In this inaugural episode we talk with Mike Tuszynski, a former CTO and current AWS Solutions Architect, along with Amit Narayanan, an AWS Solutions Architect Leader. We talk about the different roles SAs play in helping businesses use technology to successfully implement their business strategy. Our guests also share thoughts on what types of roles best prepare you for becoming a Solutions Architect and vice versa.AWS Hosts:Nolan ChenMalini Chatterjee