theCUBE + NYSE Wired: AI for CFOs & AI Leaders in Silicon Valley | Rajesh Rao, Lilt

Clips
More from theCUBE + NYSE Wired: AI for CFOs & AI Leaders in Silicon Valley

Rajesh Rao

COO & CFO

Lilt

play_circle_outline Connecting Silicon Valley and New York through theCUBE and NYSE Wired community

play_circle_outline Importance of CFOs in driving efficiency and integrating AI into business

play_circle_outline Collaboration and face-to-face interactions for trust and innovation

play_circle_outline Tech-Forward CFOs Spearheading Industry Transformation: Bridging Silicon Valley with New York Tech and Finance

play_circle_outline Business model transformation in AI era driving structural changes

Info
Transcript

Rajesh Rao, Lilt

Rajesh Rao

COO & CFO Lilt

TheCUBE is hosting a special event in Silicon Valley with NYSE Wired community, connecting experts in AI from Silicon Valley and New York. Rajesh Rao, CFO of Lilt, co-founded the initiative to discuss how AI can transform businesses, especially for CFOs. The event aims to bring real-world experiences and practical AI applications to the forefront, emphasizing the importance of driving efficiency and integrating AI into core business strategies. Rajesh highlights the significance of multilingual capabilities in AI and the need for customization to ensure relia... Read more

explore Keep Exploring

What is being discussed about the Silicon Valley office and studio, infrastructure, and connections with the New York Stock Exchange on the show floor? add

What is Rajesh Rao's position at Lilt? add

What is the significance of face-to-face collaboration in events compared to digital and virtual events? add

What is the bridge between the worlds of money and finance with technology that the tech-forward CFOs bring? add

What is the significance of business model transformation in the current era and the impact of rapidly advancing technology on the way business is conducted and value is consumed by customers? add

bolt Powered by CUBE AI

Rajesh Rao, Lilt

search

>> Hello, everyone. Welcome back to theCUBE Special presentation here in our Palo Alto Studios in Silicon Valley. I'm John Furrier, host of theCUBE. We are doing theCUBE plus the NYSE Wired community Special digital twin event. Of course, this is our Silicon Valley office and studio and infrastructure, as well as our studio in the New York Stock Exchange on the show floor where we're connecting Silicon Valley and New York. You'll see content here in the valley with local folks, experts and leaders and AI as well as in New York, and connect them with the open network that is theCUBE plus NYSE Wired community. Again, super new and growing very, very fast. And this week, we're doing the AI for CFO Summit plus our AI innovators in Silicon Valley. Of course, since we're here, the co-founder of this initiative, I'd call co-founder because he was one of the key people behind it, is Rajesh Rao, chief financial officer of Lilt. Rajesh, you are the key to success on why we're here. But well, we live here, by this cube, but why we're doing the AI CFO program. So thank you and welcome back to theCUBE.

Rajesh Rao

>> Yeah, no, thank you. Excited to be back here and I think it is amazing how fast things move. And you, me and Brian started talking about the Summit, and I passionately believe that the CFOs are in a very unique position. And given what AI is going to do from a transformative powers to the business, the CFOs sit in this key position, I'm sure we'll talk about it, in terms of not only driving efficiency improvements, but also figuring out how can we actually make AI a core part of the business.

>> Rajesh, first of all, the great vision, we're going to get into it in much detail, but I want to tell the story because I think there's been a lot of debates about return to office. Events are certainly back. Hybrid with COVID was digital and virtual events, and then hybrid came out. But what was interesting about this dynamic of the NYSE Wire Community with theCUBE and now other people plugging into this network is the relationship and the collaboration is face-to-face. And you can only get that face-to-face... You can get it on Zoom and other mechanisms online if people have that trust, but face-to-face is actually the intimate resource. So when theCUBE was starting to do events last year, we started hearing conversations, this is on my side, about CFOs, not CFOs specifically, but the tech conversation. Kubernetes, microservices, cloud-native developers, shift left cybersecurity, large-scale with Amazon and all the hyperscalers, we call it super clouds, Snowflake, Databricks, everything we do, semiconductors, Broadcom, Nvidia. All that action, which is great, by the way, is causing a disruptive enablement. And the disruption is language, like the insurance policies don't match the data, and the expensive data is valued the same as the unexpensive one. That's weird. And then cybersecurity. You got agents, so risk management. So a lot of these conversations, if you put them on a board be like, "That's the CFO's job." And so I'm like, "Okay." I meet Brian Baumann's at the NYSE and he was hearing the same thing. So when we started talking about, "Brian, the CFO conversation isn't about going public. It's not about speeds and feeds on the balance sheet and the income statement and the cashflow, or counting the beans or doing this, that and the other thing, and managing the compliance. That's the job. But it's really architecting the business transformation." We're like, "Oh yeah. We should do something." You came on theCUBE at the NYSE Wired event that we had, digital twin event in New York a couple weeks ago, and Brian was riffing with you. Now he's talking to you. You said that's a great idea. You chime in, and then it all came together like just a meal. It's just, "Oh wow. That's magic." Take us through the conversation you had with Brian Baumann because I think this was a key point because we were like, "We're doing it." That was two weeks ago, and we now have two days of programming. Great reception in Silicon Valley. Great buzz in New York. We're going to be there in January. What did you say to Brian that made him go, "We're doing this?"

Rajesh Rao

>> Absolutely, yeah. I'll call out what he exactly said, which is I don't think it ever would have been possible if it was a Zoom conversation. So we were talking about it and I was just riffing about. And this has been key, has been weighing on me on how do we get this really important conversation, which seems sometimes just in the CTO and the technology space, but it has massive implications in the business. I can feel it in my job, and we are an AI company. So obviously, I'm trying to convince other CFOs to think about from this perspective, but I just noticed there's an amazing platform that you have built. And by the way, what you have built here is phenomenal from a data-forward, tech-forward media platform. And as we were riffing about it, and I just said, "I think I've been trying to figure out a forum to do it." And then he's like, "Well, John and I have been discussing about it," and next thing you know, we are brainstorming and let's make it happen. And I think it amazes me is I think this is the power of bringing the Silicon Valley mindset of take an idea to fruition and action in two weeks. And the next thing you know, we had this summit set up.

>> And then you get consensus. You have other peers, and then we're in an era of, I call, social media 1.0, where it's like, okay, it's the same old hot boring takes or they're warm takes or cold takes, and a lot of fake data, and then with real experiences, but work's getting done now. So now you're talking about work product, whether it's media, whether it's finance, whether it's exchange work that Brian does or someone's trying to do their job. At the end of the day, the business model is grounded in the architecture of the venture itself, which is the product, the financial strategy and the money to support the ability to sell it.

Rajesh Rao

>> Yeah, yeah. Totally. And I think why bringing peers together, people who are actually dealing with real problems and facing these decisions is important is you can get a theoretical construct of like, oh, this is how finance and the CFO should be thinking about it. It should be the CTOs. But yesterday evening, when we got a whole bunch of these CFOs together and we had conversations going, the richness of that dialogue around, "Hey, I'm trying to figure out how to bring my compute costs down. How am I trying to get insurance costs to a place where I'm not paying for crazy amount of insurance because of my data." The peer-to-peer conversations, I think the quality, the richness and the nuance on that goes to a totally different level. And that's what I think someone like me, I don't want to read general analyst reports. I want to actually hear it from people who've actually done it.

>> Yeah, and when I met with Brian, just how theCUBE got involved was he had an event, similar reception here. They do these things where people are face-to-face, and they're big name people, not on the major thought leader circuit, but they are thought leaders in their domain. And I'm like, "Damn, this is really, really good. How do we replicate that?" That became like, "Okay, let's just do a digital twin of theCUBE. We'll come into the studio and we'll have those same conversations and we'll record them and share them in the open. No one owns it. Just push it out," and the magic happens. And this is where the connectedness comes in, and this is the power of what AI is going to do. And I think the face-to-face digital twin thing with theCUBE, the NYSE, these nodes and the network so to speak, the smart nodes and the network effect shows where agents are going. Now, one of the hot technical areas in the conversation now is graph databases. A network is just a graph of connections that could form at any time to do something, and AI will operationalize that. This is the power of the new environment. What's your reaction to that?

Rajesh Rao

>> 100%. And I want to make a quick shout out, John, to what you guys have built here. It's one thing to build a media platform, but I think the richness of the data and the predictive analytics and the ability to mine in a very multimodal way, I have not seen it. It's really, really impressive, and I think you're on the cutting edge of that. And combining that with human experiences and people with expertise getting on that, this is gold in terms of content. So congratulations. .

>> Well, I appreciate it. I appreciate this. It's all about the people involved and again, I appreciate it coming from you because your business is at the forefront of this revolution because you're doing things to extend this democratization, the scalable data model.

Rajesh Rao

>> Yeah.

>> So coming from you, I appreciate it. It's a huge compliment. Let's talk about your business, because this is something that fascinates me because large language models, LLMs, so everyone's heard that term, or SLM, small language model, they call them foundation models or multimodal for computer vision. Let's take the language models. It's in English, most of it. You're doing stuff in other languages, and this is where you're starting to see the global tsunami of value being created because it's just the tip of the iceberg. Now it's expanding in multiple languages. This is what you guys are doing. There's an interview, if you're watching this, there's an interview on YouTube with Rajesh and I in New York for our New York City version of the subnet content previously where we talked a lot about the company. But I want you to take a minute to explain real quick what you guys do because what you're working on is a data-driven approach. It is enabling value and sets the table for the automation, the agentic, the business model transformations around the world.

Rajesh Rao

>> Yes.

>> .

Rajesh Rao

>> Love it. Love it. So what is Lilt? So we are an enterprise multilingual platform that offers a whole bunch of different capabilities. I'll talk to two of them that I'm really excited about. One of them is exactly what you said, which is LLMs are primarily built on English, and I strongly believe that generative AI has now the ability to democratize all kinds of capabilities, whether it's education down to the masses, whether it's medical care, whether it's legal services, and I think that's happening. Unfortunately, a lot of what the LLMs have been built today is primarily in English, and you now need to make them multilingual. For that, you need AI expertise. You need multilingual data. You actually need a human in the loop for reinforced learning with the human feedback, and Lilt is uniquely positioned to do that. So we offer that. We are helping companies that do large language models to make them multilingual. And it's a very collaborative exercise to do it. So that's one of our key capabilities. And as you can imagine, it's one of our fastest growing capabilities. The other capability is providing very high quality translation that sits in for enterprises across different platforms. We have the infrastructure to go grab the content, translate it. Again, we have our own homegrown language models that we use with human in the loop to provide high quality translation at a fraction of the price.

>> Obviously, theCUBE was envious of the language because we have English, and I don't speak those other languages. We have audiences all over the world, but we're just one of many businesses in this. And I want you to, it's a very nuanced point, but I want to address it because I think it's important. The idea of language in a sovereign land or country where they speak that language, that's one thing. Then there's the English or American company or any company that is dealing in a multinational way, has language challenges. So it's not just learning the language of a group. It's knowing that when you're doing business process, there's also multilingual things going on there too. This is a huge point because sometimes, systems seem to, would decouple that and be like, "Okay. We got language covered here. Translate, localize," and that stuff's been around for a while, but you're getting at something that's a little bit different because AI can now work seamlessly across processes and across languages. So you're a multi-language model.

Rajesh Rao

>> Yes.

>> A multi-large language model.

Rajesh Rao

>> Yes.

>> Talk about that distinction because I want to make sure people understand the nuances there. It's obvious when you talk about it, but it's process driven too.

Rajesh Rao

>> Yeah. So I think when you get, as you said, when you get into the business processes, so you have a software platform now that has to dynamically adjust to different kind of languages depending on the users that are coming in. It's not just a static change in translation. So how do you embed, in the code, the translation capabilities that are, again, high quality, they have to be in context, have the right jargon. You cannot mix up, and the connotations can be completely different if you get it wrong. So getting all of that inbuilt, and if you don't have those capabilities, is going to be really hard for you to expand into different countries. I'll give an example. Some of our customers are life sciences companies, and there, they're talking about drug trials that have to go through regulatory approval through multiple countries, and the stakes are really, really high. You cannot mess that up, and there's no way you'll get drug approval across Europe unless you have had all those, the language content. And so I think what we do is really hyper-customization. One of our biggest superpowers is we have our own, as you said, multilingual large language models that train on your data. And as you get more data, it's continuously learning, and that's the name of the game. You have to train it for your local domain. You have to have domain expertise and you need customization.

>> Rajesh, you have a good point because from whether it's a startup or a regulated large public company or global company, it's the same answer. I want it to work. So you hear about practical AI from Andy Jassy at AWS when he was giving the keynote with Matt Garvin just a couple of weeks ago. We have CFOs coming on the program here saying real ROI. The word real is always in the conversation.

Rajesh Rao

>> Yes.

>> I had an interview with the CIO of JPMorgan Chase, Lori Beer. She came on theCUBE. She's a CIO, but she's got a $17 million budget. They move $10 trillion a day in business. So pretty significant. She said it just can't not fail. It has to work 100% of the time. They have zero tolerance for failure. Now in the gen AI world, she can't run the, "Well, it's going to iterate." No, no, no, no. It can never fail. The stakes are too high, to your point. This is where the practicality becomes big. How do you talk about that piece one, and two, how do other CFOs manage their AppSec review or their framework, resilience framework, whatever they're running, to ensure that it'll work? Because they have to get... Every company that's global has multiple languages, so they have to deal with that.

Rajesh Rao

>> Yeah. So great points, and I think it's really comes down to risk management. So if I'm a CFO for a very large company or a C-level executive, I cannot take a risk on a new technology that has a potential risk of creating huge negative impacts. And that's one of the challenges right now with general LLMs, is that they hallucinate. The reliability is not there once you take it and try to customize it to a local application. So the way we think about it, and that's where we, I think, bring value, is you actually bring in domain expertise. You bring in agents that are really trained on your local data. And so this is where customization really is important. And at the end of the day, and still today, I think the human in the loop is the last point of preventing failures. The trick is really figuring out how do you drive down all the benefits, the efficiencies, while maintaining the quality? And that's where the secret sauce comes in. You bring those two together. And that's a great point about ROI, John. I, as a CFO, I'm constantly pushing two things. One is I need you guys to drive and adopt AI very aggressively in every function. At the same time, the proposals you're bringing, they had to be vetted to the level of detail so that I can feel the ROI is real because I think it's pretty easy to get caught up at the surface level. And my mantra is that's why I think the CFOs cannot be a surface level. I just understand technology at this level. You have to go deep to really understand it.

>> Yeah. And the CFO, go back to any business school one-on-one conversation, the business model of a company and the strategy of what they do, why they do it, and how they do it, and how they measure it is a financial function too. And the financial strategy has to match the business strategy. The business architecture and the financial architecture have to be aligned 100%. And so the CFOs have a unique view. They are the blueprints. They know the skeletons of the organization relative to process and money flows. So when it comes to identifying, tweaking, or refactoring the business model, they pretty much have the playbook. They have the blueprints. They should be leading the business model transformation. What's your reaction to that?

Rajesh Rao

>> I think that is very much the genesis of why we started talking about this and this forum exists, which is I strongly believe, I think the CFOs have a unique vantage point into the business. They understand both the business dynamics and how it shows up, reflects in the economics of the business, the finances of the business. And I think it's their responsibility to think about how do you deploy capital in the most efficient way so that you get the business outcomes. And so I think to that point, I think they have to be driving the transformation. They have to educate the board about it. They have to lead them through it, and there is going to be some change management required, but they have to play a very critical role in that.

>> Rajesh, great to have you on the team and congratulations on all your success as an innovator in AI and certainly, as a CFO. Love your mindset and thanks for being the co-founder of our initiative and participating and being the catalyst, and being the one that convinced us that we got to do this because I think the response was great last night at reception-

Rajesh Rao

>> It was fantastic....

>> of the digital twin event, has been great, and I think we're going to do more. What do we do next? Got to do it in New York, right?

Rajesh Rao

>> Yes, absolutely. So one of the advantages of doing it here is that all the CFOs and the executives we talked to end up being very tech-forward. They're actually already adapting... Yesterday, the conversation was really, really rich when we talked to other CFOs. I believe that AI is going to transform every industry. There's no industry that's going to be unimpacted. So the more we get CFOs from even the mainstream industries, not just tech, I think we're going to get a much richer conversation there.

>> The tech-forward CFOs is a great way to call it. I was talking with Laura Peroni last night who is at Solona. She's multi-venture-backed CFO. I know if I'm to start a growing company. She was the CFO at Mist, which was a big acquisition that's powering Juniper. She's a techie. She's like, "I'm not just a CFO. I'm a techie." And that's why I like the valley vibe here. And now you bring that to New York, you've got the bridge between two worlds connecting.

Rajesh Rao

>> Yes.

>> Money and finance with tech.

Rajesh Rao

>> Yeah. No. Absolutely. Absolutely.

>> Perfect opportunity.

Rajesh Rao

>> Yeah.

>> Rajesh, thank you so much.

Rajesh Rao

>> Yes.

>> We're having a lot of fun. And again, appreciate you and looking forward to our next event in New York or online, digital twin, engagement, LinkedIn.

Rajesh Rao

>> It was a pleasure talking to you, John, and I just want to, one call out is that talking to someone from the media who has a computer science degree and has a depth of knowledge and technology is mind blowing. So I think you're the perfect person to drive this conversation and bring the thought leadership together.

>> I appreciate that, and from you, that's a great compliment. Thanks so much. Of course, we love doing what we do. We can talk all day on tech. We'd love the CFO angle. It's a great one. Again, a lot of changes coming. Business model transformation will probably be the defining moment in this era that we're in. They used to say dog years, seven years for every year. We've got AI years. Every year is 20 years. So a lot of movement's happening, but really, the game changing, structural nature of business, how business is conducted and how customers are served and how they're going to consume value is really on the table. This is the heart of the conversation. Thanks for watching. Again, more coverage here in theCUBE Studios in Silicon Valley. We have the studio in NYSE, connecting Wall Street with technology here in the valley. I'm John Furrier, host of theCUBE. Thanks for watching.