Bedrock Brief 03 Sep 2025

Bedrock Brief 03 Sep 2025

Welcome back, AI aficionados and cloud computing connoisseurs! This week in the wild world of AWS, it's all about visual search, talent wars, and a dash of good old-fashioned African expansion.

Amazon's throwing its hat into the visual search ring with Lens Live, an AI-powered shopping tool that lets you point your phone at real-world objects and instantly see matching products. It's like having a personal shopper who can read your mind (and your camera feed). But while Amazon's busy playing "I Spy" with your smartphone, they're facing a different kind of challenge behind the scenes. An internal document reveals the tech giant is struggling to compete in the AI talent war, citing issues with compensation, location, and perceived lag in the AI space. Who knew that door desks and "egalitarian" pay structures wouldn't be enough to lure top AI minds?

Meanwhile, Snowflake's expanding its AI Data Cloud to Africa via AWS in Cape Town, proving that sometimes you need to go south to move your stock price north. The move helped Snowflake buck the trend of a broader tech sector slump, with shares jumping over 22% last week. It seems that in the world of cloud computing, sometimes it pays to be fashionably late to the party – as long as you bring some cool AI tricks with you.

Fresh Cut

  • Amazon Bedrock expands to Asia Pacific (Jakarta), allowing developers in the region to easily build generative AI applications using various foundation models through a single API, with built-in security and privacy features. Read announcement →
  • AWS releases open-source CLI scripts for 60+ services, helping developers quickly create and manage resources with error handling and cleanup, using AI-assisted generation to save time and reduce mistakes. Read announcement →
  • AWS enables detailed cost tracking for GPU and AI accelerator-powered containers in Amazon EKS, helping developers accurately allocate expenses for machine learning workloads across teams and projects. Read announcement →
  • Amazon Neptune integrates with Zep, allowing developers to build AI applications with long-term memory and advanced reasoning capabilities using graph databases. Read announcement →
  • AWS Transform for VMware enables smoother migration of VMware workloads to AWS by allowing VPC CIDR range modifications, preventing IP conflicts and offering flexible IP address management options. Read announcement →
  • Amazon Bedrock's new simplified cache management for Claude models automatically reuses cached content, reducing manual effort and token usage for developers building multi-turn workflows and research assistants. Read announcement →
  • SageMaker's new account-agnostic project profiles allow developers to create and reuse machine learning project templates across multiple AWS accounts and regions, simplifying governance and accelerating setup for large-scale ML environments. Read announcement →
  • Easily analyze your Google Sheets data using Amazon QuickSight by directly importing sheets into SPICE datasets, simplifying data visualization for spreadsheet users. Read announcement →
  • Amazon QuickSight, a scalable Business Intelligence service for creating interactive dashboards without infrastructure management, is now accessible in Israel and the United Arab Emirates, expanding its availability to 25 global regions. Read announcement →
  • Amazon Q Developer now lets admins control Model Context Protocol servers for their organization, enhancing security and resource management across all supported IDE plugins and the CLI. Read announcement →

The Quarry

Natural language-based database analytics with Amazon Nova

Amazon Nova brings the power of large language models to database analytics, allowing users to query structured data using natural language. This AI-powered system breaks down complex queries into explicit reasoning steps, employing validation loops to catch errors and refine results. By bridging the gap between human intent and database schemas, Nova not only simplifies data exploration but also introduces a layer of self-correction that could significantly reduce query errors and improve data accessibility for non-technical users. Read blog →

More posts:


Core Sample

How Bayer Pharmaceuticals accelerates drug innovation with Amazon Bedrock

Bayer Pharmaceuticals is harnessing Amazon Bedrock and AWS security services to supercharge their drug innovation pipeline, combining multiple AI models with robust data protection. Their platform approach uses Amazon Macie to safeguard sensitive information while leveraging generative AI to accelerate research and development. This clever mashup of cutting-edge AI and stringent security measures aims to fast-track novel drug discoveries, potentially shaving years off the typically glacial pace of pharmaceutical innovation. Watch video →

More videos: