What I’m watching in AWS right now and why we need to keep up.
Every year AWS rolls out new features, but this year’s announcements feel different. It’s not just the typical small tweaks or iterations; the updates we’re seeing represent a real shift in capability, and they’ll influence how we build, innovate, and compete.
If you work in technology or in any industry that depends on data, these changes matter. Here are a few that stood out to me and why I think they deserve our attention.
S3 has always been the foundation for object storage, but now AWS is transforming it into something much bigger with Amazon S3 Vectors—purpose-built storage for AI agents, AI inference, and semantic search of S3 content.
For me, this is one of the most interesting announcements because it addresses a major challenge for companies working with AI: vector storage. Until now, teams used Amazon RDS for PostgreSQL with pgvector, or another solution, to implement vector databases. That added cost, complexity, and friction.
By building vector search directly into S3, AWS makes it easier to store and query embeddings with sub-second performance and up to 90% lower cost. If you’ve been exploring generative AI, this fundamentally changes your architecture decisions. And honestly, it sets a new baseline for what “cloud-native AI” means.
AI agents have been one of the most hyped concepts in tech this year, but the reality is that most companies struggle to move from prototypes to production. Security, identity, state management, and monitoring are the usual roadblocks.
That’s why the services within Bedrock AgentCore caught my attention. This is AWS addressing the problem head-on with a modular framework for runtime, memory, identity, and observability. It feels like the missing link between early experimentation and operational reality.
For Stratus10, this is especially exciting. We’re already thinking about how to leverage AgentCore services in our own Kalos platform to make it smarter, more proactive, and even more effective at delivering cost and security insights.
Healthcare and life sciences have been slower to adopt cloud innovation due to complexity and regulation. AWS is clearly committed to addressing industry barriers, and this year’s updates reinforce the trend.
HealthLake, while technically not brand new, is an important and growing service I’d recommend with just-announced enhanced compliance and interoperability. AWS HealthLake enables healthcare, pharmaceutical, and insurance organizations to store, transform, and analyze their raw clinical data in FHIR standards, and then extract information for health analysis, predictive modeling, patient care, and more.
HealthOmics, designed specifically to store, query and analyze genomics and biological data, received significant enhancements this year too, including workflow versioning, parameter automation, elastic throughput, and support for NVIDIA L4/L40S GPUs for high-performance genomics.
If you’re in healthcare, biotech, pharma, research, or related, these updates are worth a close look.
We should also be aware of the impacts of these services beyond healthcare. Because it signals where cloud is going: deeper and domain-specific with capabilities that solve real problems in industries with unique needs.
What I see across all of these announcements is a clear shift. AWS isn’t just shipping more services. They’re embedding intelligence, automation, and industry awareness into the foundation of the cloud.
That’s exciting because it opens the door to building better solutions without starting from scratch every time. For companies, it means completely rethinking what’s possible in terms of efficiency, scale, and innovation.
I’ll be diving deeper into these capabilities in the coming weeks and looking for ways to apply them in meaningful ways. If you’re experimenting with any of these, I’d love to hear what you’re seeing.
Which of these AWS updates do you think will make the biggest impact this year?
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.