I am here at AWS re:Invent 2025 in Las Vegas, and one announcement is cutting through the noise!!!!

AWS has named 60 partners in new Agentic AI categories within its AWS AI Competency. Under the hood, this is about separating signal from noise and recognizing partners that already run real, production workloads with AI agents on AWS.

This is not just another badge. It is a map for anyone who wants to move from AI demos to systems that people inside the business actually trust.

How AWS is structuring the Agentic AI landscape

From a builder’s point of view, a few things stand out.

1. Clear buckets for what you are buying

The new Agentic AI competency groups partners into three types:

• Applications
  When you want something that already works in the real world. These are products with agents built into them, handling workflows like support, analytics, code review, or sales operations.

• Tools
  When you are building your own agents and need help with data access, security, or connecting into your existing systems. Think runtimes, context engines, and guardrails that sit inside your AWS environment.

• Consulting services
  When you are rolling out agents across teams or critical systems and need help with architecture, governance, and operations.

This mirrors how most internal teams actually work:
Some things you buy, some you assemble, and some you bring in experts for.

2. Higher bar for “ready for production”

Getting into these categories is not about clever messaging. Partners have to show:

• Real customers

• Real workloads running on AWS

• How they handle security, permissions, failures, and human oversight

For customers, that means you can start from a basic level of maturity and spend your time on fit: your data, your compliance needs, your integration paths.

3. Aligned with how you actually discover and deploy

AI agents on AWS Marketplace are growing fast, and Marketplace itself is changing to match how people want to buy AI: solution first, faster deals, and smoother deployment flows. The Agentic AI competency sits on top of that and shows you which partners AWS has already vetted for serious agent work.

So you get two layers working together:

• Marketplace to discover, evaluate, and deploy

• Agentic AI competency to tell you who has passed a higher bar

AWS has also shared three key pages that are worth a close look. You can use these as a starting point to explore what is now possible.

1. Marketplace feature launch: discovery to deployment

Marketplace Feature Launch Blog:

This blog explains how AWS Marketplace is being tuned for the AI era:

• Smarter discovery, including AI driven search

• More solution based offers, not just single products

• Faster private offers and smoother deployment paths

For teams, this means less time on contracts and glue work, and more time getting agents into real workflows.

2. AI agents and tools catalog

AI Agents and Tools on AWS Marketplace:

This page is a focused catalog for AI agents and tools. It groups offerings into:

• Software that already has agents embedded

• Pre built AI agents you can adopt quickly

• Agent tools like context, knowledge bases, and guardrails

• Agent development and infrastructure solutions

• Professional services for strategy and rollout

Instead of hunting across dozens of pages, you have a single place to:

• Browse agent solutions

• Compare types of offerings

• Filter by use case and delivery method

• Keep discovery, purchase, and deployment inside your AWS environment

3. Agentic AI Competency blog

Agentic AI Competency Blog:
This blog introduces the new Agentic AI categories and lists the 60 partners that have met the bar. It also explains how this work connects to AWS services designed to run agents at scale in production.

Read this to understand:

• Which partners AWS has already validated for serious agent workloads

• How AWS is thinking about responsible, governed agent systems

• Where to start if you want to stay close to AWS native patterns

A simple 12 month playbook

If you put all of this together, you get a clear way to plan your next 12 months with agents on AWS:

1. Use off the shelf agent applications for quick wins
   Start with validated applications from the competency list and the AI agents and tools page for use cases like support, analytics, sales operations, or coding help. You do not need to start from zero for every problem.

2. Use validated tools when you need flexibility
   When you have custom data, strong security needs, or deep integration requirements, pick tools from the AI agents and tools catalog and from among the Agentic AI partners. You keep control, but you are not rebuilding the foundation yourself.

3. Use consulting partners for core or regulated systems
   Once you touch core systems or regulated data, involve consulting partners that have passed the Agentic AI competency bar. They have already dealt with governance, monitoring, and failure modes in real projects.

This is how you move from isolated demos to a portfolio of agent systems that people inside the business can actually trust.

These 60 partners are an early signal of who is already on that path.

What I will be watching at re:Invent

As I walk the floor and talk to these partners, I will be looking at:

• How they are using the new Marketplace features in real customer deals

• What kinds of agent use cases are live in production today

• How customers are balancing off the shelf applications, tools, and consulting help

I will share more conversations and concrete examples as the week goes on.

🔍 Stay Ahead in AI & Data! Join 137K+ Data & AI professionals who stay updated with the latest trends, insights, and innovations.

📢 Want to sponsor or support this newsletter? Reach out and let's collaborate! 🚀

Best,

Ravit Jain

Founder & Host of The Ravit Show