Andreas Welsch: 0:30

Today, we'll talk about how you can build your first AI agents and who better to talk about it than someone who's actively working on that with a lot of customers in Europe, Eduardo Ordax. Hey, Eduardo, thank you so much for joining.

Eduardo Ordax: 0:42

Thanks. Thanks so much for inviting me today. It's really a pleasure to be here.

Andreas Welsch: 0:47

Hey, Eduardo, I've been seeing a lot of you on LinkedIn over the last one or two years, I see you're picking up a lot of good momentum, so I'm super excited that you're spending this time, not just with me, but with the audience as well. But for those of you who don't know you, maybe you can introduce yourself real quick and tell us a little bit about yourself, who you are and what you do.

Eduardo Ordax: 1:05

Yeah, sure. My name is Eduardo. I'm living right now in Spain out of Madrid. I'm working as a Generative AI lead for AWS, managing EMEA, Europe, Middle East and Africa. And that means essentially like how we can help our customers to leverage all the potential of generative AI as well. This is what I'm doing. I'm having so much fun. I'm having so much work during the last few months. And yeah, that's me.

Andreas Welsch: 1:39

Eduardo, should we play a little game to kick things off? Yeah, sure. Let's go. Wonderful. All right. This one's called In Your Own Words. And when I hit the buzzer, you'll see the wheels spinning. And when they stop you'll see a word. And I would love for you to answer with the first thing that comes to mind and why in your own words. And to make it even more interesting, you'll only have 60 seconds for you to answer. Are you ready?

Eduardo Ordax: 2:04

I'll try. I'll do my best.

Andreas Welsch: 2:07

Good. So here we go. If AI were a vehicle, what would it be? 60 seconds on the clock.

Eduardo Ordax: 2:15

Wow. That was, super easy. I will say it will be a Cybertruck.

Andreas Welsch: 2:21

Okay. Why? I don't know. First time I saw it, it was like like something that it was coming from the future. And, many of the things that we are seeing right now related to artificial intelligence, it seems we are, experiencing some stuff that it really comes from the future, probably two or three years ago, we could not even imagine, all the capabilities, the large language models, but also any other foundation models. And I think it's the first time where we really believe like all the things that they're going to come in the next I don't know, five to 10 years, they're going to be being like a more impactful for all of us. That's why, I see the Cybertruck and I imagine like a remote world in the future, maybe on Mars. I don't know. And I'm thinking of this kind of artificial intelligence idea, whatever. So it's, as simple as that. It reminds me the cyber truck. Thank you so much for sharing. I'm curious what you think in the audience, what comes to mind now? There many ways that we can think about the, future and, the futuristic technology. Some of them are already here, like the cyber truck that is on the roads in the U.S. In many ways, I also see the parallels to AI and AI agents. Certainly it's the big hype topic of this year. I think there's still a lot of more momentum that this technology will gain, and vendors that keep pushing the technology. But I'm curious, in your work with customers, what do you see, what departments are exploring AI agents at the moment?

Eduardo Ordax: 4:02

In my opinion, it's even more than departments where they are all the customers in general, right? But even if you are asking me like which kind of customers they are more willing to start like exploring and start implementing agents, these are probably the ones who have started as well, in the very early days with AI And like all the customers from very high related industries like FSI, financial service industries insurance, healthcare, life science, because they have invest a lot of money, a lot of time, a lot of resources in terms of like also in terms of like teams and people, they have gained this kind of like competitive advantage in regards to work with artificial intelligence, right? Because They really know how they have to structure the data, their systems, and most of them they are already moving to the cloud because at the end of the day if you want to use AI, it's much better when you're in the cloud, right? Probably these customers are the ones that I'm seeing that it's been much easier for them to start implementing agents and for many different use cases, right? But most of them the most typical use case right now, it's to automate internal processes because at the end of the day you have a control over this kind of workflows, right? Because there is a lot of hype around agents and sometimes an agent, it's nothing else than a cool workflow, right? I used to call it like a cool workflow because most of the times it's just an LLM who is like making different API calls, or even doing this kind of like a self reflection. But it's like that. Like internal processes, because the thing that you can do with internal processes is you can automate like things that is taking a lot of time. It's taking a lot of like manual steps. And you are going to be more efficient and you are going to get a lot of, saving on, on, on costs. Of course, there are like customer facing for like customer support where you can, I don't know you can implement a chatbot where you are not expecting only to provide information to your end customers, but also to take actions at the end, right? Hey I want to book a room for next week in Toronto in a hotel for four for people. You are expecting this chatbot, not only to give me information about the room, but also to make the call. to the system, where I can book the room, I can give you the special offers, so I need to connect through a different database, I can retrieve the data, I can send it to you. But right now, at least what I'm seeing in Europe right now, it's mainly for internal processes like, hey, how can I automate the process to book the holidays? Or how can I automate the process, to request a specific service? Or how can I automate To open an internal ticket, to submit any, thing, whatever, right? So this is what I'm seeing the most right now here.

Andreas Welsch: 7:06

Thank you for sharing. I think that's really good perspective to also see where are companies looking at these things right now. And it seems that it's similar to AI and machine learning, RPA, and those previous hypes, right? Look for something where, like you said, it's your internal process. You know how it works. You ideally have it documented. There are ways to make it more efficient, faster, and all the like. Now that brings me to my next question that I've been thinking about. And that's does it really matter what's your first agent that you build? And are there specific departments? Is it mainly the IT department that again looks to tease out the last 20 percent of, Hey, can you reset my password? Or I forgot where my nearest printer is, or I have some other standard problems. Is it departments that are looking at this first? Or do you see business, finance,

Eduardo Ordax: 8:01

Yeah, I'm seeing a lot of IT departments, especially we're managing the life cycle of different weekends, different claims, and so on, but also like the business sales marketing departments, like how you can automate. The relationship with the customer, they are like super interested on this because as I said like, when we started to talk about Generative AI. Probably the top use case was about implementing chatbots, right? To automate the relationship with the customer, but right now I'm seeing all these departments like, hey, I want to take one step further, not only to provide information to my customers, but ultimately to take actions, right? But also something that I'm seeing a lot is as part of RAG use cases, Retrieval Augmented Generation. Like one of the challenges with RAG is as soon as you increment or as soon as you scale the number of documents that you want to query, the different techniques become so challenging, right? Because it's not easy to find exactly what you want to query. So you need to place where to find the exactly answer to your query, right? So we are seeing this kind of like a gigantic rack where essentially it's again, the same thing is okay, we are taking different approaches I don't know, hybrid search, query rewriting, query ranker implemented through agents, right? So this is very effective because I can initiate different journeys. Just to make sure that I'm going to retrieve the most important part of the most important tag for my previous question. Agentic RAG is also one of the top use cases that probably I'm seeing right now. Not because of agents, but because of RAG that like it's, probably one of the things that most of the customers, they are implementing. There is nothing else than a knowledge search, right?

Andreas Welsch: 10:03

I think you mentioned a good point about RAG documents different approaches to RAG and we see so many schematics of agents and their components in, our LinkedIn feed in, newsfeeds day in, day out, right from planning to memory core modules, everything else. What are the, key technology components that you actually need if you want to build an agent or use an agent?

Eduardo Ordax: 10:28

At the end of the day like an agent, it's nothing else than an LLM doing some kind of action, right? So if we look into an LLM like, the most simple part, let's say I'm going to ask something to the LLM and it's going to give me an answer, right? An agent is okay, I'm trying to take one step more. So I can do this kind of specific planning okay, if I'm asking you to write a document, I'm not going to write a document in one thought, but I'm going to say okay, I need to analyze what I, need to do, this kind of cell reflection. So first of all, I'm going to come up with different answers. I'm going to analyze if my answers that are good or not. So it's like the model. It's asking itself okay, how good, how bad this is going to be. And I'm going to make a plan, right? Okay, I'm going to do first this, second this, then this. But also you can use external tools, right? So it's an LLM, as I said okay, you can start a specific action. So your LLM, because they have all of these LLMs, most of them, they have what they call this function calling or tool use. There is nothing else that, hey, me as an LLM, I can make an API call to an external system and I can retrieve this information. So if I'm asking you for some information that I don't know, I can say okay let's go through this website. I'm going to find the answer and I'm going to amend the answer that I'm going to provide with this information. This is the part of the agents that you need is like the LLM, where you can write this, kind of like a self reflection where you can analyze. So it's pretty much like us, right? If I'm asking you to write something, you're not going to do it in one suit. Maybe you will write a document, you will do a draft. And even before a draft, you're going to start like planning okay. This is the schema of my document. Okay. This is the schema of my document. I start a first draft and you're going to say okay, I'm going to review my draft. I will review it and I will say it's good. It's bad if it's not good enough. Okay. I'm going to take another review, right? So these are just LLMs. It's not like responding at the first time, but doing this kind of unicellular reflection, analysis, planning. And at the end, even if I need it like I can make different calls to different tools. And if we take it in one step more, sorry, this is pretty much like us, right? Like we are not an experts on everything, right? So at the end of the day, we will have like many agents working together, right? So it's not only one agent to do all different steps, but it's okay, I'm going to put working different agents on different tasks. So you may have an agent that is going to be an expert on finding information in the website. Okay. This other agent is going to be the expert trying to orchestrate all the others. Another expert is going to be the one who is going to book your travel to Toronto. So you will have many different agents, but the thing is okay, I need even to orchestrate those agents, right? That it's called this meta agents and multi agent collaboration. There are like many different names, but I will say these are the four main aspects like self reflection, planning, tool use or function calling and collaboration across agents.

Andreas Welsch: 13:49

Now we've, been talking in the tech industry for a long time about microservices, have individual capabilities compartmentalized so we can again, call them or string them to together. That definitely sounds a lot like that. And even on a more granular level. But I'm curious too for someone who might be new to this topic. If you hear all about agents in the news, do I have to build all of this myself? Do I have to either be a developer or do I need a team of developers? Are there some things where I should put or use things off the shelf? What's your recommendation between maybe off the shelf and building this yourself? When does it make sense to build it?

Eduardo Ordax: 14:29

Yeah, I think at the end it depends, right? There are many different approaches and there are many different flavors, right? If you have something that is super specific that is Hey, I want to do this, that is going to be A, B and C. You have already like many different agents that they are going to integrate with your ERP. Your different systems and you don't need to create it from scratch, right? Hey, you can use the agents from Salesforce, or you can use the agents for created virtual assistant, or you can create the customer support agents. They're like many different companies that they are creating these agents. And then removing all this heavy lifting about how to manage the agents, how to manage the resources, how to query different databases. So if you don't have the expertise, if you don't have the skills, you Maybe it's better to go through this approach, right? But they're like all the different flavors, right? With AWS, you have agents on Bedrock. It's still you can customize which kind of agents you are going to build, right? You can say, okay, I want to build these specific agents for customer support. You are going to have access to all these different systems. But again, it's going to be a managed service. It's much easier to implement. You can give some instructions. You can build different workflows. So even like the entry barrier is going to be very low, right? But even let's say you have like very specific requirements. You want to create everything from scratch. It's you have frameworks like I dunno, land graph, right? So with LangGraph, you can create everything from scratch. Probably the most complicated part is like the level of abstraction is so high and the entry barrier is high as well, right? So it's not probably the best way to go for everyone in the business unless you have enough experience, right? You have different approaches based on your skills, based on your capabilities. If you build just like these kind of vertical agents, okay sales agents or HR agents entry barrier are very low, but the capabilities or the like possibilities to customize these agents, they're not so high, right? Because it's like very specific. You have something in between like using agents on Bedrock, where you have many things to customize. Essentially, you can customize almost everything, but still, you build on top of a framework that it's already defined and it's there. And then you can build everything from scratch, but you need to manage your agents, you need to manage your resources, you need to do this level of abstraction. See you on SOAP. Again, it depends. It depends on the customer. I've seen all different aspects and I think it's not different from like a normal software. Like probably you are going to use SAP, so don't build it from scratch, but if you need to do something very specific, you will have to do it, right? So it's, not different from, Traditional software, to be honest.

Andreas Welsch: 17:18

That makes a lot of sense, and I think that's good reassurance, right? Look at what your standard vendors already put out there. If you need something else or more than that, or highly customized to your, not just industry, but business to build it, on top of the platform. Now, I'm curious, in your work how far advanced do you see companies be on this journey? Are they just scratching the surface and trying to figure out what are these agents? Are they any good? Where can we use them? Or do you already see, leaders think about how do we make sure that they act in, in the same way? Maybe that they use the same tone the, same style how they respond, that they are grounded in, the same documents, maybe? Not just the information, not just the business documents, but a code of conduct or values or say if they're a finance agent that they're grounded in IFRS accounting standards. How far along is that thinking there from what you're seeing?

Eduardo Ordax: 18:18

So I think I've seen a shift over the last few months, right? Of course, Europe is different from the U. S. In the U. S. they are like much more advanced right now, right? But what I've seen during the last probably 12 months back. It was mainly about experimentation, even they were trying to build something like a POC that it was going into production, but it was like, hey, we're implementing agents, but the scope of these agents, it was super limited, right? So it was like agents working in production, but with a very limited scope. Let's say If I want to implement an HR agent to book your holidays or whatever, it was exposed only to like a reduced number of employees, for example, right? What I'm seeing right now is this is starting to change. So it's not only like to expose these agents to this kind of control group, but trying to expose it to, everyone, right? So of course here, the main challenge is about the outcomes, it's about the cost. But right now I'm seeing the customers trying to move more like to this use of agents. It's in general, right? It's agents and it's AI at the scale, because what I think is to be honest, at the end of the day, everything is going to be around agents. But because if we think about generative AI, just about like LLMs, I think it's useless, right? Let me explain, right? It's not like it's useless, but we as humans, we are expecting not only generate to provide or to receive information, but at the end of the day, it's about accomplish a specific task, right? And this is what we are getting from by using agents. That's why I'm saying like LLMs, they are useless because at the end of the day, all these companies that are expecting to automate processes, right? So it's not Hey, I'm expecting to know which is the HR policy, for vacation. But ultimately to book my holidays on the, tool through agents, right? So what I'm seeing right now, it's more and more customers that are implementing agents at the scale. Of course it depends on the industry digital native customers, startups like customers that they don't have a legacy, an IT legacy for them is much easier, right? It's more natural because even they don't have the problem about the data. They usually have already the data lake, everything is integrated and so on. But also customers from hybrid related industries FSI, insurance, healthcare, life science. They are really implementing agents for many different like use cases, like I don't know research discovery that it could be like a super complicated. You can use agents to do that. So I think within the next 12 months, we are going to see an explosion of your like people consuming these services. And I think that's good because this is the life cycle, or this is the flywheel that, that we really need because as soon as we have more people consuming these services, we will get much better models, much more capable models, because at the end of the day, you like to train these models, you need a lot of, money, right? So all these companies like OpenAI and Anthropic. They need to be profitable. They need to make a business with that. So as soon as we see more people using these systems at the scale, we will see a clear improvement of these LLMs and foundation models as well.

Andreas Welsch: 21:40

I like how practical you make that and how you share what you're seeing. I think it's great to see companies experimenting with this, but also looking at what are the next steps. Things that, that we should be doing and, connecting what you said with our earlier point of use what your out of the box vendors offer, combine it with something that's more customized on your platform. Build it there. I'm just wondering with scaling the number of agents and even if there are many very specialized tasks, you put them together in a collaborative workflow and in a collaborative setting, how do they, communicate? How do we make sure that they communicate well, that they exchange the right information, that they expect the right input, give you back the right output? We've established, protocols, right? Things like TCP iP, HTTP, things like that in the past that handled that communication. Do you think we'll see something like that to ensure interoperability going forward as well? Do we need that?

Eduardo Ordax: 22:38

Yeah you made the 1 million question, right? Hey, how these models, they are going or how these agents, they are going to communicate. I still find people that they believe that they are like autonomous, super intelligent systems. that they will know how to communicate across many others to accomplish tasks, right? And we are not there yet. Like these agents, they are not people, right? But what we can do is we can organize these agents pretty much like us, right? Where we can have different kinds of organizations. We can have this kind of like master agent that is going to do like many different things, but we can have like hierarchical agents where we can define, okay, this is going to be the master, right? And these are going to be their employees. So I can assign different tasks for different agents. So I just need to make sure that I establish a protocol or relationship between these agents. This is much, much easier, right? Because it's okay, one single agent connecting through many different, right? But it's only like one to one connection across all these, agents and you have control, but there is only one kind of connection, right? Or one kind of protocol, but you may have as well, like higher levels of, hierarchies, like you may have One agent, then you have different departments and within the different departments, you're going to have a specialist agent. So you are including or you are adding one layer more of complexity, but even you can have peer to peer agents where we can work as a team, but there is not like one single agent that is going to act as a planner, but we can work all together at the same time. So to do that, there is a manner to do that, right? Like even you can implement that, as I said before, like by using like a land graph, but you need to define first this kind of relationships, and you need to be very clear on the structures, like every single agent they are going to accomplish, and also the parameters and the different like variables that they are going to use between others. And as I said within Amazon Bedrock we recently announced to reinvent this kind of like multi collaboration agent. Where you can define these relationships very easily, where you can say okay, this is going to be my like a master agent, and it's going to communicate with A, B, and C. It's going to send all these different parameters. It's going to expect all these different answers. I'm going to allow them to call these different systems. So like we are doing all these different like a heavy lifting behind the scenes. But for you, it's going to be much easier, right? But again, you need to define the relationships between the agents. Otherwise like you can expect something that, you won't like it. If you just leave them to their own, that's why, yeah, it's super important. And maybe in, in the near future, We will see something more closer to like autonomous agents where they can, you know, interact and so on but for sure, we are not there yet.

Andreas Welsch: 25:43

Okay. Sounds like it's a journey, but many have already embarked on it. Now we're getting close to the end of the show and Eduardo, I was wondering if you can summarize the key three takeaways for our audience today.

Eduardo Ordax: 25:54

I think first of all I will say about trying to remove all the hype about agents try to see what is the value, right? And the value of an agent, it could be as simple as, hey, I just need to build a simple workflow. That is going to do A, B, and C. That could be an agent. So don't worry too much about all this cap, autonomous agents and so on. Let's try to start by the basics. My recommendation is to start also like with tools that they're going to allow you to move faster. Like probably at some point of time, you will be in a situation where you can build everything from scratch, right? But at the beginning it's about, trying, testing, experiment, fail, and then try it again. My recommendation is to start with tools that are going to reduce this time to market. And I think like tools like Bedrock or even Vertical Agents it's going to help you a lot on the process, right? Probably what I would say the third element is to be very clear on identify those processes where you are going to get a lot of value, right? Like Agents can be super powerful, But it's not about overcomplicating the stuff, right? It's about trying to identify those processes that right now seems like super manual, where you are losing a lot of money hey, I need to do this every single day, or I'm not even able to automate it, whatever. Trying to identify all these different parts of your processes that they are lacking a lot of automation. I'm trying to see how like it will be by using agents and so on. So that will be my three takeaways. Try to remove all the hype on agents. Sometimes it's just pure workflows. Second, start by experimenting a lot and try to reduce the time to market. Use platforms that are going to allow you to move faster. And third, again, it's all about, platform development. Finding the value where you're going to implement it. So try to identify the steps of your processes that right now don't look pretty well and try to automate it by using agents.

Andreas Welsch: 28:02

Now, that sounds like really, sound advice. And I know it's, grounded in the experience that you have, and that you see every day working with customers in Europe. So Eduardo, thank you so much for joining us and for sharing your experience with us today.

Eduardo Ordax: 28:16

Thanks. Thanks a lot, Andreas, for inviting me today.