Multicloud AI & ERP / Cloud Migration, Integration & AI Operations
Why it matters
Keeping up with releases is a full-time job, but the right new service can remove months of custom engineering. This summary highlights what matters for enterprise teams building on AWS.
How it works
The big themes
Agents are the headline: Bedrock AgentCore went generally available in October 2025, Nova Act is GA for browser agents and Kiro replaces the Q Developer IDE experience. Data for AI got cheaper with S3 Vectors, GA in December 2025. Silicon advanced with Trainium3, and operations gained the AWS DevOps Agent, GA on March 31, 2026.
At a glance
| Agents and models | Operations | Data and cost | |
|---|---|---|---|
| Headline release | Bedrock AgentCore (GA Oct 2025) | AWS DevOps Agent (GA Mar 31, 2026) | Amazon S3 Vectors (GA Dec 2025) |
| Models and chips | Nova 2 family, Nova Forge | European Sovereign Cloud (GA Jan 15, 2026) | Trainium3 |
| Developer tools | Kiro, frontier agents | Q Developer plugins (supported until 2027) | SageMaker Unified Studio |
| Retired or changing | Q Developer IDE to Kiro | App Mesh (retired Sep 30, 2026) | Q Business and Kendra to Amazon Quick |
What stands out
Agents and models
Bedrock AgentCore gives agents a managed runtime, memory, gateway, identity and observability; Nova 2 brings Lite, Pro, Sonic for voice and Omni for multimodal; and Nova Forge lets organizations build custom Nova models.
Operations
The AWS DevOps Agent investigates incidents across telemetry, the European Sovereign Cloud offers EU-only operations and Amazon Quick brings agents and BI to every employee as Q Business and Kendra close to new customers.
Data and cost
S3 Vectors cuts the cost of large vector indexes, S3 Tables manage Apache Iceberg, Trainium3 lowers training and inference cost, and Bedrock Guardrails can now protect models outside Bedrock.
Design decisions that matter
- 01Pilot AgentCore for new agents
Instead of building runtimes yourself.
- 02Move large vector sets to S3 Vectors
Where latency allows.
- 03Plan App Mesh and Q Developer moves
Before support ends.
- 04Benchmark Trainium3
For training and inference cost.
Common mistakes we see
- Ignoring retirements. App Mesh is already gone.
- Adopting everything at once. Pilot one service per quarter.
- Building what AWS now manages. Custom agent runtimes and vector stores.
A practical 30-day plan
- Week 1
Check your estate for retiring services.
- Week 2
Pilot AgentCore with one agent.
- Week 3
Test S3 Vectors for a large index.
- Week 4
Plan Kiro and Quick adoption.
How NTS helps
NTS keeps AWS clients current, running architecture reviews against new releases, pilots of services such as AgentCore and S3 Vectors, and migrations off retiring services before deadlines.
What you get
- A briefing on the latest AWS AI services for your business
- Use-case fit assessment for Bedrock, AgentCore and Kiro
- Pilot projects on the newest capabilities
- An adoption roadmap
Typical outcomes
- Early, safe adoption of new AWS AI services
- Clear priorities for AI investment
- Competitive advantage from new capabilities
How we work with you
- 1Assess
A focused review of your AWS AI adoption across AWS, Azure and Google Cloud. We look for the gaps we find most often: ignoring retirements, adopting everything at once and building what AWS now manages.
- 2Design
A target architecture and roadmap built on the decisions that matter most: pilot AgentCore for new agents, move large vector sets to S3 Vectors, plan App Mesh and Q Developer moves and benchmark Trainium3.
- 3Build
Hands-on delivery alongside your team. We check your estate for retiring services, pilot AgentCore with one agent, test S3 Vectors for a large index, and finally plan Kiro and Quick adoption.
- 4Run and improve
We brief your team every quarter on new AWS AI services and update your roadmap.
Why teams choose NTS for AWS AI adoption
- Always up to date
We track every AWS AI launch so you do not have to.
- Hype filtered out
We tell you what is ready for production and what is not.
- Hands-on proof
Pilots show real value before investment.
Book a free 30-minute review
Talk to an NTS architect about your AWS AI adoption on AWS, Azure and Google Cloud.