Yong Quan

Excited to announce that Amazon Bedrock AgentCore is now generally available! This comprehensive agentic platform is set to transform how organizations move AI agents from prototype to production faster. Customers and partners frequently share that while AI agents show incredible promise, deploying them reliably at enterprise scale requires robust infrastructure and controls. AgentCore is our answer to this challenge. The response since preview has been phenomenal - with 1M+ downloads of the AgentCore SDK, adoption across industries, and internal implementation by teams like Amazon Devices Supply Chain and Operations With AgentCore, you can use any framework or model without compromising security and with standards such as MCP and A2A supported out of the box. It represents a fundamental step forward in how you’ll deploy, operate, and scale AI agents. If you're not exploring how to leverage agents in your organization, you risk being left behind as others lean in and innovate. Curious to learn more? Check out the blog for all the details: https://lnkd.in/gVqkT4dh Let's shape the future of AI agents together. The possibilities are incredible!

693

5 Comments

People often ask me about the differences in architecture between Amazon Dynamo (the 2007 SOSP paper), DynamoDB (the AWS serverless NoSQL database), and Aurora DSQL (the AWS serverless SQL databases). I captured the key points on durability, high availability, programming model, and trade-offs in my blog: https://lnkd.in/gkMG_tVD TL;DR: Despite the shared name, Dynamo and DynamoDB have little in common architecturally, and make very different trade-off decisions.

631

8 Comments

I am always excited to hear about customers's journeys on how they use Google Cloud Databases innovations to transform the experience of their users. With state-of-the-art AI models, Agentic AI (MCP), AI retrieval, AI Query Engine, AlloyDB, a 100% PostgreSQL database, is becoming the platform for customers to re-engineer their applications and systems for AI. In this blog by Vishal Vaibhav and Melissa Ludack, Target shares how they partnered with Google, leveraging AlloyDB's industry-leading AI capabilities to reimagine their search architecture to bring structured data (SQL) & unstructured data (text, vector search) to overhaul their search bar! Target has a sophisticated system for search, with AI models that enable nuanced semantic matching using AlloyDB's ScaNN index technology. It is a great testament of the research at Google in AI (ScaNN is the underlying tech used by Google Search and YouTube, for example) to see it also powering retail search on Target.com and helping their users! Congratulations to Target and look forward to partnering with you to bring further exciting advances in AI in GCP Databases to delight your customers! #Partnership #Databases #RetailInnovation #TargetSearch #GoogleCloud Lots of great insight in their blog on how Target leverages AlloyDB AI to create personalized, delightful shopping moments for their customers every day. https://lnkd.in/gaGPNZD5

151

7 Comments

AWS transform was featured in the Amazon Q2 quarterly earnings release as "General availability of AWS Transform, an AWS agent that cuts mainframe modernization timelines from years to months, completes VMware network conversions up to 80x faster, and reduces operating costs for .NET applications by up to 40%." See https://lnkd.in/gSrsSC72.

85

You can now read and write to a Unity Catalog table from a Serverless EMR compute engine via the UC open API. And yes, it supports both Delta and Iceberg. Read more here- https://lnkd.in/eyc3VGCQ #unitycatalog #databricks

114

3 Comments

This is a great example of the real-world impact of AI capabilities within AlloyDB for PostgreSQL ! Target set out to modernize its digital search experience to better match guest expectations and support more intuitive discovery across millions of products. To meet that challenge, they rebuilt their platform with hybrid search powered by filtered vector queries and AlloyDB AI. The result: a faster, smarter, more resilient search experience that’s already improved product discovery relevance by 20% and delivered measurable gains in performance and guest satisfaction. #alloydb #postgresql #gcp

49

Is #kubernetes the right substrate for AI? 🤔 Every day I hear from developers and companies talking about the complexity of operating #AI workloads on Kubernetes. There are strengths - a mature API and storage/networking primitives - but modern AI applications require scaling beyond the datacenter. Customers complain about spanning their kubernetes workloads across global regions, dense YAML configurations, and a complex stack to maintain. At Runpod - we have tens of thousands of enterprise #nvidia & #AMD #GPU's, spread across 28 global data centers. Each day, we have dozens of customers who go viral - their inference rapidly scaling across our global fleet automatically. Customers want the flexibility to run on-demand 2 hour training runs and have a B200 #Slurm cluster ready in one click. Replicating that with #k8s is achievable, but it requires a substantial investment in traffic routing and cluster management. Most developers and teams I have spoken to don't want to deal with the overhead - they want to focus on delighting their customers. Kubernetes will always have a place in large internal clouds and corporations - however these teams are also demanding novel solutions which take advantage of a world of compute at their fingertips. 🏎️ Thank you for the conversation on the AI Infra Summit stage, Hyungjun Kim, Meeta Dash, and Aaron Lasota. Ruby Goodall and the AI Infra Summit team hosted an incredible event - it is always amazing seeing the innovation happening across the industry.

104

3 Comments

🚀 Amazon Managed Service for Apache Flink now supports Customer Managed Keys (CMKs), giving you enhanced control over your data security. With this new release, you can leverage your own AWS Key Management Service (KMS) keys to manage encryption for your streaming applications—helping organizations meet stringent compliance, governance, and auditing requirements. You now can exercise granular control over data encryption at rest, align with security best practices and regulatory needs, and confidently process and analyze streaming data at scale. By integrating CMK support, AWS continues to provide innovative, flexible, and secure solutions for real-time stream processing. https://lnkd.in/gMEBynSJ #ApacheFlink #AWS #Data #streamprocessing

61

2 Comments