Clips
News
More from Boomi World 2024

Michael Bachman

Head of Research

Boomi Innovation

Harnessing AI product integrations: Boomi’s strategy for modern application development

If one looks at the generative artificial intelligence movement today, it’s all about next-gen applications and AI product integrations.Those applications are either being retrofitted, repurposed or refactored to net new applications being built in the open-source development community, according to theCUBE Research executive analyst John Furrier. There’s very robust demand from developers for gen AI, and Boomi LP is well-positioned given its installed base workflow and data access.“Boomi has the keys to the kingdom,” Furrier said. “They have the access to the jewels of all the top enterprises, and they’re in a good position to make that work

play_circle_outline AI integration at Boomi World

play_circle_outline Practical AI and its applications

play_circle_outline Customer stories and success with practical AI

Info

Michael Bachman | Boomi

Michael Bachman

Head of Research Boomi Innovation

>> Welcome back, everyone, to Boomi World. I'm John Furrier, host of theCUBE. We got Savannah Peterson out on the ground on the show floor, getting all the action from the partners and the players here at Boomi World who are making AI integration happen. Savannah, over to you.

>> Thanks, John. It's so great to be here in beautiful Denver, Colorado at Boomi World 2024. My name's Savannah Peterson and you're watching theCUBE. And I'm very excited to be chatting with someone who leads research and strategy on the emerging tech side, on the innovation team. I'm here with Michael. Thank you so much for joining us tonight.

Michael Gold

>> Glad to be here. It's a great night, isn't it?

>> I know we've got the DJ pumping right now. I can already... There's people dancing. A lot of good energy in the Boomi community.

Michael Gold

>> A lot of energy, and there should be, too. We've got a lot of really exciting things happening, not only as announcements, but what we're doing practically for everybody on the planet right now.

>> Casual statements, Mick.

Michael Gold

>> Casual statement.

>> Can you give us a couple of those exciting things?

Michael Gold

>> Yeah. Well, I'm not going to steal too much thunder away from the big guy when he comes on stage, but what I will say is this: when it comes to artificial intelligence, I want people to really understand that Boomi's not only in the game, we're going to lead the game from where we're at.

>> about how you're going to do that. We're talking practical AI, correct?

Michael Gold

>> Practical AI. Yeah. Our strategy involves two different ways that we're going about it. One is AI in the platform, and one is using the platform to create good AI related out promise. That's where we're focused for practical AI today. It's an awesome set of stories. It's an awesome set of technology and capabilities. Yeah, we're excited about it.

>> Can you give us some examples of how people are using practical AI at Boomi?

Michael Gold

>> Sure. By now, everybody knows about large language models. Probably everybody's used ChatGPT or some analog of ChatGPT, but the model underneath ChatGPT can be exposed in a useful way, and those models aren't really trained on data sets that are private, right?

>> Right.

Michael Gold

>> So it's important to be able to train those models safely and securely and accurately, or to use those models safely, securely, and accurately with your own data.

>> I bet some people are a little nervous about that.

Michael Gold

>> They are indeed. However, when you're talking about practical implementations of artificial intelligence, generative AI, this type of thing can be augmented with something called retrieval augmented generation. And so that's a mouthful.

>> Let's talk about RAG.

Michael Gold

>> Yeah, let's talk about RAG. That's how we go with it.

>> What does that mean?

Michael Gold

>> So basically what RAG means is a way to use your own data to make the LLM hallucinate less, if at all. And if you have the right set of capabilities within a platform to be able to make AI not only, well, less hallucinogenic, but actually grounded, and you can provide safety wrappers around everything.

>> They need to be grounded as an AI.

Michael Gold

>> So the opposite of hallucination is grounded, and one aside on that, Boomi actually means ground or earth in Hindi. And so it really fits into the way our platform was conceived back in the day.

>> Full circle ethos there. That's nice.

Michael Gold

>> Well, and it's funny you say that too, because that full circle is really what... We didn't intend for large language models to actually be here 20 years after the founding of Boomi, yet here we are. And now more than ever, our platform with all of its capabilities actually make LLMs super useful, super practical, and businesses can actually put them into production, like a couple of customers that are here today that have used Boomi to do so.

>> We're really at a juncture where it's a lot of hype versus actually making AI real. Love that you wanted to talk about practical AI. I know you've got some great customer stories that you're able to share. Can you give us a little taste of one of those?

Michael Gold

>> Sure can. We have customers speaking about this here at Boomi World '24, but one of the things I'll say is everybody really would love it if chatbots didn't... Well, I won't say didn't suck, I just said it.

>> You can say whatever you want.

Michael Gold

>> Fine. We want chatbots that don't suck, and actually we're at a point in history where we can make sure that chatbots are actually good, useful, and can serve people well. One way that chatbots are useful is with data from your public website. We have customers that are actually using RAG-based approaches to LLMs to incorporate their data so the LLM doesn't hallucinate, and customers no longer need to do keyword search lookups and that sort of thing. Or a chatbot has an inadequate response to whatever the query is going to be. Now, you can just talk in plain-speak in your native language.

>> NLP, baby.

Michael Gold

>> NLP, exactly. And you can query in your natural language and get answers that you want.

>> Have you ever had an experience with a chatbot that was positive?

Michael Gold

>> Yes, I have.

>> That you didn't build?

Michael Gold

>> That I didn't build? Okay. One. I have a couple buddies. There's a gentleman on our team who built this amazing framework for making interactive chatbots through voice commands.

>> Oh, nice.

Michael Gold

>> And so he interviews one of the LLMs. Well, I decided to do this at a conference last week, so I run this AI conference in Philly. We actually tailored that conversation around this particular chatbot in front of hundreds of people. Complete demo, it did everything. We asked existential questions.

>> And there was no scary twist?

Michael Gold

>> There was no scary twist. People got a little bit concerned, but we gave the large language model a voice. We gave her the ability to interact with the crowd, so they were able to ask questions in real time and she was able to respond. And she was a she, because we gave her a voice and she identified as an entity that was more female. Anyway, so that was-

>> That's awesome.

Michael Gold

>> It was awesome. So to answer your question, the crowd loved it. They loved it. And we didn't get for her to name herself, but that was the next step. Anyway, I've done that too, and it's pretty amazing. Anyway, that is not usually a production level use case that we would talk about in a business context, but you asked the question, have you ever had a chatbot that didn't suck? And she was awesome.

>> That's a perfect example.

Michael Gold

>> Example of that. And then the customer that implemented the chatbot that they're now using on their website, it too does not suck. It's really good. It's retrieving information about products that they have, and giving that back to their customers, and it's a super useful use case, not only for them or other companies within their industry, but it's a ubiquitous type of use case that anybody can apply. And by the time folks leave Boomi World, they're going to know that Boomi is able to help out with these types of use cases and many others.

>> Sure. I love that. The energy in the room definitely reflects that. All of your partners are really delighted to be your partners. They all want to be your favorite , everyone I've talked to you. You have Over 800 partner here, 20,000 customers.

Michael Gold

>> Isn't that amazing?

>> It is amazing. And I bet... What does this week mean to you, then?

Michael Gold

>> Yeah, it's a coming out party of sorts. We've reinvented what it means to be iPaaS. We've reinvented, or we're about to create the new era of what it means to have integration and automation. And so the next layer above that is orchestration and beyond. And so the idea is if you're going to work with this awesome engine called a large language model, generative AI, even discriminative AI like our traditional machine learning, it is important to make sure you get the data to that particular model in the right pipelines. We provide all of the capability to be able to build those pipelines so that you can actually work with the LLM in a way that's meaningful to the business. And we're there. We are this orchestration framework, and that is why we're so different. And I'm not even looking at our legacy competition any longer. I'm totally switching our gaze onto the future because that's what our customers are demanding. They don't want us to fight over table scraps of where we were. They want us to really focus on what the needs are in the future with this technology that's real. And if we can make it real for them now, they're going to become fans overwhelming. That is what we're going for.

>> Yeah. Well, I think you're succeeding, at least from the folks that I've talked to. Any goals, any partners you're particularly excited to catch up with this week? It's got to be like a family reunion.

Michael Gold

>> It really is. It's our first time here since 2019 with all of these partners.

>> The before times.

Michael Gold

>> The before times, yes, for the word we shall not name, for the illness that we shall not talk about. But that said, I'm really excited to see all of our partners and customers that made the journey to be here. And even the ones that couldn't be here, they're just as excited. You see my bow tie? It's from Australia. I met with a ton of partners and a ton of customers down in Australia just a couple of weeks ago. Also, Singapore and Malaysia. Shout out to APJ. Anyway-

>> Is the climate similar there?

Michael Gold

>> It is. These folks are hungry for this technology, and to make it meaningful, useful, practical, and safe. Also, responsible is good. Speaking of responsibility, there's one other thing I want to wax poetic on.

>> Please.

Michael Gold

>> We are going to be making a couple of big announcements around responsibility, and so we want to set really the stage for what's to come with responsible use of AI going forward.

>> When can we expect this to happen?

Michael Gold

>> That will be coming out in the next couple of days.

>> Great.

Michael Gold

>> Yeah, yeah, yeah. So actually, by tomorrow we're going to, actually tomorrow you'll hear more about .

>> Thanks for whetting our whistle. Michael, that's a bit of a tease. Last question for you, since we talked about AI, but you also look at a lot of emerging technologies given your role. What excites you the most about the future?

Michael Gold

>> What excites me most about the future is in spite of all the hype, is really I am net positive when it comes to what we have for the future. I'm excited that we can look at this new alien technology and make it useful with a variety of different emerging domains, in a variety of different industries. IoT can become real. API management can become real. Real new real use cases that are going to be governance layer that we've never had to have before that we can provide. I'm excited about IoT in ways that I couldn't have been prior to these two types of AI, discriminative and generative AI. And the fact that we can orchestrate all of that together is really exciting to me. What I'm really excited about, though, is the innovation that the customers are going to be able to do with platforms like ours. And so I am really excited to see how folks are going to build the future.

>> And we're excited that you're able to help empower that future, and also feel more confident in embarking on that journey with partners like you. Michael, thank you so much. It was great.

Michael Gold

>> It was a pleasure.

>> Practical AI, RAGs, we're talking about governance. We really covered all the bases. And bow ties, no less.

Michael Gold

>> Bow ties, too.

>> I look forward to seeing what ties to bring to the rest of the conference today.

Michael Gold

>> You will see plenty of bow ties.

>> Good.

Michael Gold

>> It's all good.

>> I appreciate it. And thank all of you for tuning in wherever you are on beautiful planet Earth today. We're here in Denver, Colorado at Boomi World 2024. My name is Savannah Peterson. You're watching theCUBE, the leading source for enterprise tech news.

Michael Bachman | Boomi

Michael Gold

>> Glad to be here. It's a great night, isn't it?

>> I know we've got the DJ pumping right now. I can already... There's people dancing. A lot of good energy in the Boomi community.

Michael Gold

>> A lot of energy, and there should be, too. We've got a lot of really exciting things happening, not only as announcements, but what we're doing practically for everybody on the planet right now.

>> Casual statements, Mick.

Michael Gold

>> Casual statement.

>> Can you give us a couple of those exciting things?

Michael Gold

>> about how you're going to do that. We're talking practical AI, correct?

Michael Gold

>> Can you give us some examples of how people are using practical AI at Boomi?

Michael Gold

>> Right.

Michael Gold

>> So it's important to be able to train those models safely and securely and accurately, or to use those models safely, securely, and accurately with your own data.

>> I bet some people are a little nervous about that.

Michael Gold

>> Let's talk about RAG.

Michael Gold

>> Yeah, let's talk about RAG. That's how we go with it.

>> What does that mean?

Michael Gold

>> They need to be grounded as an AI.

Michael Gold

>> So the opposite of hallucination is grounded, and one aside on that, Boomi actually means ground or earth in Hindi. And so it really fits into the way our platform was conceived back in the day.

>> Full circle ethos there. That's nice.

Michael Gold

>> You can say whatever you want.

Michael Gold

>> NLP, baby.

Michael Gold

>> NLP, exactly. And you can query in your natural language and get answers that you want.

>> Have you ever had an experience with a chatbot that was positive?

Michael Gold

>> Yes, I have.

>> That you didn't build?

Michael Gold

>> That I didn't build? Okay. One. I have a couple buddies. There's a gentleman on our team who built this amazing framework for making interactive chatbots through voice commands.

>> Oh, nice.

Michael Gold

>> And there was no scary twist?

Michael Gold

>> That's awesome.

Michael Gold

>> That's a perfect example.

Michael Gold

>> Isn't that amazing?

>> It is amazing. And I bet... What does this week mean to you, then?

Michael Gold

>> It really is. It's our first time here since 2019 with all of these partners.

>> The before times.

Michael Gold

>> Is the climate similar there?

Michael Gold

>> Please.

Michael Gold

>> We are going to be making a couple of big announcements around responsibility, and so we want to set really the stage for what's to come with responsible use of AI going forward.

>> When can we expect this to happen?

Michael Gold

>> That will be coming out in the next couple of days.

>> Great.

Michael Gold

>> Yeah, yeah, yeah. So actually, by tomorrow we're going to, actually tomorrow you'll hear more about .

Michael Gold

>> And we're excited that you're able to help empower that future, and also feel more confident in embarking on that journey with partners like you. Michael, thank you so much. It was great.

Michael Gold

>> It was a pleasure.

>> Practical AI, RAGs, we're talking about governance. We really covered all the bases. And bow ties, no less.

Michael Gold

>> Bow ties, too.

>> I look forward to seeing what ties to bring to the rest of the conference today.

Michael Gold

>> You will see plenty of bow ties.

>> Good.

Michael Gold

>> It's all good.