🦄 Making great presentations more accessible.
This project enhances multilingual accessibility and discoverability while preserving the original content. Detailed transcriptions and keyframes capture the nuances and technical insights that convey the full value of each session.
Note: A comprehensive list of re:Invent 2025 transcribed articles is available in this Spreadsheet!
Overview
đź“– AWS re:Invent 2025 - The Zero Migration Path from Data to Enterprise Agentic AI (AIM123)
In this video, Trianz presents their Concierto platform's zero-migration approach to enterprise agentic AI. The platform offers five solutions: modernization, migration, management, maximize (observability), and insights/agentic AI. Key challenges addressed include the 18-24 month traditional data platform deployment versus the 11-month industry transformation cycle. Concierto enables federated data connectivity across on-premises, cloud, and SaaS environments without moving data, productizes data based on usage rather than storage, operates on zero-code principles, and leverages existing governance. The DALAI assessment evaluates agentic AI readiness across data domains. Implementation takes 90 days, with AWS funding available for assessments. Built on AWS stack with Anthropic Claude and Bedrock, it includes native Gen BI, conversational AI, and enterprise agentic marketplaces.
; This article is entirely auto-generated while preserving the original presentation content as much as possible. Please note that there may be typos or inaccuracies.
Main Part
Introducing Trianz, Concierto Platform, and Federated Data Connectivity Expertise
Good afternoon. We're going to talk about the zero migration path from data to enterprise agentic AI. It's a provocative title, but hopefully the conversation will live up to the hype. A quick word about Trianz: we are an AWS ISP partner. We work with AWS in multiple disciplines. We bring a completely different approach and a model to help our customers transform, and we call that Transformation Services as a Software Model. We're going to talk very quickly about what our approach is and what our capabilities are, and then dive into this concept of zero migration data to agentic AI. We lead with a massive platform called Concierto.
Concierto has five big transformative solutions in it. Number one is modernization—modernization of infrastructure, data, and applications all at the same time. Number two is migrate. Concierto can assess, map out dependencies, create waves, and move all three layers at the same time to any of the three clouds and between the clouds. Number three, Concierto allows you to manage all of these environments from a single pane of glass with a single abstracted process. Number four is maximize. We use the word maximize in the context of observability because what Concierto does is it looks at consumption patterns, capacity, events, and uses AI and ML to remediate anomalies so that you're able to optimize how you're using the cloud. And then the last piece is insights and agentic AI, which is what we're going to dive into. Concierto is hosted on AWS. We have over 60 to 70 customers in the last five months alone who have adopted this, not to mention a number of customers worldwide who are using the platform through Concierto Partners.
Before we talk about this concept of data to agentic AI, a quick word about our experience in all things data. I'm specifically focusing on federated data and connecting with a federated model. We've done a lot of work in this space, but what is relevant here is that about four to five years ago, we were the first company to connect the AWS cloud with data sources anywhere on the planet. AWS subsequently acquired the software from us, and we ended up becoming a very close strategic partner. We do a lot of engineering services for AWS as well. This is the foundation on which we're going to be talking about this concept of data to agentic AI without migrations.
Yesterday, AWS released about a dozen connectors to third-party ISPs from the AWS cloud. Concierto and Trianz is the company that built those connectors, and we're going to keep doing that over the course of the next several years until we are able to create, co-create, or help in the design and build out of agentic cloud. We're also working with AWS in connecting the different layers of the cloud itself. That is the experience and perspective that we bring to the table.
The Urgency of AI Transformation and a Zero-Migration Approach to Agentic AI
So let's dive into our subject today. There are two really big existential questions that everybody on the board is asking. If you're anywhere in the data landscape, there is no hiding from these two questions. Question number one: how much time do we have to get ahead of AI? It's not about getting to AI because everybody else in your industry is doing the same. Therefore, you have to put yourself in a position where you're innovating and coming up with ideas that are ahead of the industry. You can only do that when the foundation that you require to reinvent the business using AI is in place. Number two is how much time do we have to get there?
When we look at the way data platforms are created today, data platforms are modernized, purchased, and deployed, whatever the model may be. For a large enterprise, it takes a minimum of eighteen to twenty-four months to get there. Until several years ago, this approach was fine because we all had time. Now what's happening around the world is that
thanks to innovators and leaders in any industry, the change cycle—the time in which transformation of an entire industry takes place, or at least in the big spots—is now less than eleven months. No matter what industry you are in, something big is changing within eleven to twelve months at most. Therefore, business in reality is living on borrowed time. So why take an approach that is expensive, does not guarantee an outcome, and is most likely going to be obsolete by the time it is done?
The other thing that we have to keep in mind is that fundamentally, your data is screaming and saying "I don't want to move," or a large chunk of your data. Why is that? There are compliance issues due to which you have to keep some data on-premises. You have stuff sitting on SaaS. You make an acquisition, you sell a business. There are fifteen different things above data technically speaking that are always changing. We may have a majority of our data on one cloud, which is fine, but the goal is not to move the data itself. The goal is to make use of the data wherever it is.
What is going to happen—and I'm sure we all read the same news, everybody is immersed these days in understanding AI and its effects—but what I would like to suggest is that ultimately the outcome is going to be binary. There are those that will succeed and there are those that will fail. The old way of looking at the middle of the ground, the middle of the pack—I do not think that middle of the pack paradigm is going to survive for a long time. Those who get it right, obviously, will see positive outcomes. Those who do not get it right will see a very troublesome and painful pattern in front of them: slow obsolescence, customers walking away, revenues going up and down. You will come to a point where you cannot reverse the trend. We have seen enough examples in the entertainment industry, in real estate, in retail, in fashion, and it is going to happen pretty much in every single industry as we move forward.
On the other hand, those who get it right will start seeing revenue spikes because these customers who are going to leave the companies that are obsolete still need the services and products. They are going to move to those companies that appeal to them and that give them a better value and what they are looking for. So what we did specifically in this space is we tore up our playbook and we said we are going to look at everything from a clean sheet. What we set out to do is change the paradigm and build a very different approach to get from data to insights as fast as possible, as quickly as possible. Second is make it really simple for the business to be able to use. Third is make it scalable across the company.
In order for us to do this, we examined six fundamental questions that keep data leaders awake all the time. How do we enable AI on our existing assets without moving them? How do we productize data and look at data in small pieces as opposed to running expensive, time-consuming queries? How do we break the bottleneck of business to IT—where the business needs something today and the data team says they have a six-week SLA to deliver that? Business does not have time anymore. How do we break that? How do we achieve governance in large enterprises without the bureaucracy that goes along with it and the time that we lose?
Number five is how do we get people in the data universe—your architects, your data scientists, your governance folks, your security folks—they all need to be able to look at the same picture. Finally, how can we do this as fast as possible? So we artificially put a stake in the ground and said we should be able to do this in ninety days. And here is how we solve problems. Number one is we connect with data, structured and unstructured, in real time using a federated concept. We can connect with data pretty much anywhere on the planet, and if we cannot, we will be able to build it in four to six weeks. Number two is we productize data based on how it is used, not based on how it is stored.
We can do this really fast using AI. Buy your business domains, buy your data domains. Number 3, the entire platform is built on zero code. So nobody has to type any queries and do a lot of technical stuff in order to get to the results. Number 4 is we use governance that companies have built over decades at source and we enhance those governance rules as opposed to lifting and shifting all of the data and then rewriting and rebuilding the stuff.
Number 5 is that we assess everything in one shot in about 4 to 5 weeks. We are able to get everybody from data teams, including your business on the same page as to where we are and where we are going and what it really takes for us at the applications level, at the data level and infrastructure level to be ready for Agentic AI. Yes, we have been able to break the 90 day barrier.
Concierto Insights Platform: From Data Marketplace to Enterprise Agentic AI in 90 Days
So I'll now leave you with some concepts and visuals so that you can understand what this Concierto Insights and Agentic AI platform is about. Number 1 is federated data connectivity. No matter where your data is on-premises, on cloud, on SaaS, wherever it is, we are able to connect to it instantly. Number 2 is we are able to productize data and catalog it. Number 3, we publish the data in a marketplace that is very similar to Amazon Prime Video or Netflix. So it is very intuitive for business users to be able to see data just like you see movie titles by genre: sales, sales finance, sales compensation, sales ops, and so on. We organize data very intuitively, and number 4 is we have native Gen BI, native conversational AI, but you can also integrate with tools that you may have invested in already. Conversational AI and then ultimately agentic marketplaces.
Let's take a very quick look at this. So Concierto brings universal connectivity, as I said. Number 2 is we use AI to rapidly productize within hours and days your entire data the way you want to see it and the way you are going to consume it. Real time, zero copy. Concierto does not store any of your data. It sits at your source. It only brings the results from queries to generate the insights that you need. On the other hand, if you want to start migrating data and do just the incremental stuff, we can create a copy on the cloud as well. Intelligent caching is what we are talking about, and it accelerates future queries and so on. Governance, as I said, is at source. Number 2, enterprise data marketplace.
Each of these tiles that you see has a bunch of icons. What those icons are showing business and IT teams, just like a movie title user ratings or a TV show rating, is discovery lineage. You can join two data products and create a super data product and still see the entire lineage. Data quality, usage, and how ready this particular data set is for agentic stuff. Now, this exposes data to users as they want to consume it. But it also tells data teams as to what their target should be. For example, if I have a data quality problem across the enterprise, I do not have to solve it across the board. I can solve it for my high priority data sets to make sure that the results of those are much more accurate.
Now, every company today has invested, large enterprises especially, in some sort of BI capabilities. We are not here to ask companies to replace those and buy yet another tool. Concierto is automatically connected to Tableau and Power BI and AWS QuickSight and so on, but in order to truly help democratize, Concierto also has native Gen BI capabilities. It brings your cost down and it helps everybody in the company truly use data. So this is data democratization in real life. Concierto is built on the AWS stack with Anthropic Cloud, Bedrock, SageMaker, and so on. On top of that we have written thousands of rules which make your queries much more intelligent, much more precise, and very accurate.
Your business users can essentially come in and start having a conversation with their data. You don't need to create canned reports and keep feeding them every week, every three days, or whatever the periodicity is. What conversational AI does for any data set that a user has access to is automatically read through the data needed to answer the query and give you quantitative insights, charts, and even decisions that the user needs to make.
The last piece in this is enterprise agentic marketplaces. When we've gotten our applications and our data ready after seeing the scores and making improvements, Concerto has a capability where the data teams and the application teams can start designing agents and make those agents available to users to embed in their processes and automate decision making and execution.
This whole journey starts with a very simple exercise called DALAI, which stands for Automated Data and Agentic AI Readiness Assessment. The name is inspired by the Dalai Lama and is about wisdom, getting informed, and getting educated. When you deploy Concerto in your landscape, the platform will generate a report covering your data, its demographics, where it is stored physically and virtually, how it is growing, the profile of your databases, what exists, what goes out, what is end of life, your provisioning, your storage, quality, security, and all of this is laid out very neatly.
On top of that, you and our teams will work together to create a scorecard for agentic AI readiness. You give the rules and add the rules to those already in the platform. For each of the data sets by business function and by data domain, the platform will tell you whether that data is ready for agentic AI and if not, what needs to be done. Once all of this is done, we lay it out neatly in the form of a report that your business users and IT leaders can consume, and then we are off and running.
The beauty of this approach is that while we are doing the assessment, teams can work together and configure Concerto and design an agentic AI for real world use. We begin with a ninety-day exercise in which we take one business unit or one business function, create a template for you for using the platform the way I described, and in parallel we train teams so that this becomes a self-service capability inside the organization as opposed to data teams needing an SI or consulting firm to come and do all of this work for you.
While the leaders are consuming the feedback from the agentic AI and the data assessment that we do, the platform gets ready. In ninety days you'll be able to turn it on and make it available to your users. The good news in all of these assessments is that they are funded by AWS. You do not have to spend a lot of money just to get an assessment done. We and AWS are partnering to fund this for customers and deliver these reports for free. After that you can choose to continue if you like it. There's no obligation. You don't have to continue with the platform if for whatever reason you don't.
That's the offer we are bringing to the table for every customer and even partners who want to use this platform for their end customers as well. In the background, if you are truly married to moving your data to the cloud, now that we have decoupled it, Concerto can also migrate your databases, modernize them, and move the data itself to the cloud. It can create lake houses on the fly on cloud native technology as opposed to very expensive third party tools. That really is the path from data to agentic AI in ninety days.
Trianz is at booth 1381. There are a number of folks here from Trianz as well. Please stop by and let's have a conversation. Thank you very much.
; This article is entirely auto-generated using Amazon Bedrock.
Top comments (0)