Alexander Gallego, founder and Chief Executive Officer of Redpanda Data, discusses the company's recent developments and strategic positioning at the forefront of data streaming technology in theCUBE's "Mixture of Experts" series, filmed at the New York Stock Exchange.
In this insightful episode hosted by Dave Vellante, co-founder and co-Chief Executive Officer of SiliconANGLE Media Inc., Gallego shares the latest updates at Redpanda Data, highlighting its Series D funding led by Google Ventures. The conversation delves into the challenges and innovations within data streaming, as well as Redpanda's position in the market, particularly through its significant partnerships and technological advances such as the rollout of Apache Iceberg.
Key takeaways from the discussion include Redpanda's strategic milestones and growth, as highlighted by Gallego. Notably, Redpanda now powers the New York Stock Exchange's cloud data feeds, underscoring its capability in mission-critical environments. Furthermore, the conversation addresses the shift from batch processing to real-time streaming, an essential change driven by new capabilities in AI and the innovative use of data for business process improvements, which Redpanda enables, as Gallego explains.
Forgot Password
Almost there!
We just sent you a verification email. Please verify your account to gain access to
theCUBE + NYSE Wired: Mixture of Experts Series. If you don’t think you received an email check your
spam folder.
Sign in to theCUBE + NYSE Wired: Mixture of Experts Series.
In order to sign in, enter the email address you used to registered for the event. Once completed, you will receive an email with a verification link. Open this link to automatically sign into the site.
Register For theCUBE + NYSE Wired: Mixture of Experts Series
Please fill out the information below. You will recieve an email with a verification link confirming your registration. Click the link to automatically sign into the site.
You’re almost there!
We just sent you a verification email. Please click the verification button in the email. Once your email address is verified, you will have full access to all event content for theCUBE + NYSE Wired: Mixture of Experts Series.
I want my badge and interests to be visible to all attendees.
Checking this box will display your presense on the attendees list, view your profile and allow other attendees to contact you via 1-1 chat. Read the Privacy Policy. At any time, you can choose to disable this preference.
Select your Interests!
add
Upload your photo
Uploading..
OR
Connect via Twitter
Connect via Linkedin
EDIT PASSWORD
Share
Forgot Password
Almost there!
We just sent you a verification email. Please verify your account to gain access to
theCUBE + NYSE Wired: Mixture of Experts Series. If you don’t think you received an email check your
spam folder.
Sign in to theCUBE + NYSE Wired: Mixture of Experts Series.
In order to sign in, enter the email address you used to registered for the event. Once completed, you will receive an email with a verification link. Open this link to automatically sign into the site.
Sign in to gain access to theCUBE + NYSE Wired: Mixture of Experts Series
Please sign in with LinkedIn to continue to theCUBE + NYSE Wired: Mixture of Experts Series. Signing in with LinkedIn ensures a professional environment.
Are you sure you want to remove access rights for this user?
Details
Manage Access
email address
Community Invitation
Alex Gallego, Red Panda
Alexander Gallego, founder and Chief Executive Officer of Redpanda Data, discusses the company's recent developments and strategic positioning at the forefront of data streaming technology in theCUBE's "Mixture of Experts" series, filmed at the New York Stock Exchange.
In this insightful episode hosted by Dave Vellante, co-founder and co-Chief Executive Officer of SiliconANGLE Media Inc., Gallego shares the latest updates at Redpanda Data, highlighting its Series D funding led by Google Ventures. The conversation delves into the challenges and innovations within data streaming, as well as Redpanda's position in the market, particularly through its significant partnerships and technological advances such as the rollout of Apache Iceberg.
Key takeaways from the discussion include Redpanda's strategic milestones and growth, as highlighted by Gallego. Notably, Redpanda now powers the New York Stock Exchange's cloud data feeds, underscoring its capability in mission-critical environments. Furthermore, the conversation addresses the shift from batch processing to real-time streaming, an essential change driven by new capabilities in AI and the innovative use of data for business process improvements, which Redpanda enables, as Gallego explains.
In this Mixture of Experts segment from the New York Stock Exchange, theCUBE’s Dave Vellante sits down with Alex Gallego, founder & CEO of Redpanda, to unpack how open table formats, real-time streaming and AI agents are reshaping enterprise data strategies at the intersection of tech and finance. Gallego shares updates since his last appearance: Redpanda’s Apache Iceberg support is now GA from the streaming engine perspective - enabling data sent to Redpanda to be immediately queryable by engines like Snowflake, Databricks, BigQuery and Athena - and the comp...Read more
exploreKeep Exploring
What companies were involved in the fundraise and why did the board decide to invest in the company?add
What organization now powers the New York Stock Exchange cloud data feeds?add
What are the competitive advantages and unique value adds of Redpanda in the market compared to other similar technologies?add
What is the protocol that Redpanda is currently betting on for securing data access for agents with their suite of connectors?add
>> Welcome back to our Mixture of Experts series here at the New York Stock Exchange. We're overlooking the options trading floor. My name is Dave Vellante. We're super excited to have Alex Gallego here. He's the founder and CEO of Redpanda. Alex, good to see you again.
Alexander Gallego
>> Yeah, great to be back.
Dave Vellante
>> It's been nine months. You were in our studio in Palo Alto. A lot going on, a lot of rumors swirling. I know you can't comment on it, but the information in January broke a story, for those of you didn't see it, that Snowflake is looking at Redpanda as an acquisition. You guys just raised a series D, lot going on since we talked nine months ago. Give us the update.
Alexander Gallego
>> So, where we sit in the market, it's super strategic to all of the data warehousing technology. So, we just released Apache Iceberg as GA and as far as I know, sort of the first company to release this from the streaming engine perspective. And so, what it means is that, people can now send their data to Redpanda and it's immediately queryable by all of the big query engines, whether it's Snowflake or Databricks or BigQuery or Athena. And then the fundraise, led by Google Ventures with participation from Lightspeed Partners is really the same insider investors. We have a great board. They decided to step in as they see us super strategic in this shift towards agentic AI and where Redpanda sits in the market. And so, last year we closed on our record year exit, we brought in all of the large electric car companies in the U.S. as customers, we powered two out of the top five banks in the U.S. So, it was a really great year for us. And so, just super excited to be here and thanks for having me.
Dave Vellante
>> You doing anything with NYSE?
Alexander Gallego
>> Yeah. In fact, yeah, we are super proud to announce that we actually now power the New York Stock Exchange cloud data feeds. So, anytime that you're consuming data from NYSE in AWS, that's powered by Redpanda. And so, super thankful of the partnership with Sridhar here and the team at the New York Stock Exchange for launching this. I think, for them, just for context, they are 10x better than any competitor in the market. And so, just excited to be able to help them break new ground.
Dave Vellante
>> I want to go back to something you said nine months ago. You referred to Iceberg like the TCP/IP of data. You remember that?
Alexander Gallego
>> Yeah, yeah.
Dave Vellante
>> Which I thought was great. But what's happening here, for those of you don't follow is, Iceberg is an open table format and it really has come about, it's really changed the way that people think about data. I mean it was, Snowflake popularized, okay, we're going to put data in the Snowflake cloud and they'll take care of everything, and that was awesome, separating compute from storage. And then you saw what Databricks did, is they sort of popularized any compute engine operating on any data. So, separating data from any compute engine, and that's what you just referred to, you can now operate, bring any engine to the data and you're enabling that.
Alexander Gallego
>> Yeah, exactly. Well, so what happened is, remember that those of us that grew up during the Hadoop era, if you remember those eras-
Dave Vellante
>> Yeah, you have some scar tissues from that.
Alexander Gallego
>> It's the 3:00 AM pager, that's how you know. We didn't want to manage those systems. And so, we wanted a SaaS product that would give us infinite scale and we got that, but at the same time, we ended up jailing our data. You're like, okay, I can now hire a company and I get infinite scale, but in order to query the data, I have to use that particular API and that particular render. And so, what Iceberg did is like, okay, let's take back ownership of the data and now I can bring any query engine and if I'm using a particular query engine, it's because it's delivering the most value to the company right now. And in the future, by the way, you still have your data, so you can bring any query engine. And so, just, we, I think Redpanda's accelerating the growth and adoption really for the world's largest companies of Iceberg.
Dave Vellante
>> So, you've talked in the past about the world and mentioned Hadoop world's moving from batch to real time. And you said you consider yourself ring zero in streaming in the data pipeline. Can you explain that transition and how it's going to affect the world?
Alexander Gallego
>> Yeah, so there are two tailwinds. There's the batch to real time, and then obviously the big elephant in the room is AI, right? It's just like, how's that change? So, let's talk about first, the batch to real time. It turns out, it's easier to model the world as it's happening event by event, rather than creating this artificial barriers of saying like, okay, I'm going to process all of my events at midnight or at 6:00 AM or at noon. The world doesn't happen in this nicely delineated events. It doesn't make any sense, but we invented that as a way to deal with complexity. So, that was batch. And then what people realized is that, with real time streaming, you can actually just make more money, if you... An example I like to give is credit card withdrawals at ATM. So we had a customer, they were a bank, they were a batch company, and they would send you emails, oh, hey, would you like to sign up for additional credit? Great. They moved to micro-batching and they now do it every hour, and they saw some improvement on NYFT. When they moved it at the time, in which you insert your credit card or debit card at the ATM, would you like to use this withdrawal as a credit? They saw a 300% lift in conversions. And so, that's the power that real time can have on an existing business. And then there's obviously AI and it would be weird to submit a query to OpenAI now and get a result at midnight, because you forget what the question was.
Dave Vellante
>> So, I remember when we were talking, you, John and I. A while ago, we were talking about fraud, and I was like, "Yeah, fraud detection's pretty good now." But you gave a great example around gaming, where people are committing fraud with the fake money, which you can monetize, and that's a very hard problem. Can you explain that use case? And maybe there are other examples that aren't. I mean, your point was, okay, if somebody's trying to make a transaction from Nigeria and you're in the States, that's easy, geolocation, you can do that. But the examples that you gave are much, much more challenging. And so, I'm curious as to what kind of use cases that will open up.
Alexander Gallego
>> So, let's talk about the gaming. And so, with gaming, we power the world's largest gaming engines today. So, think like, hundreds of thousands of users, 50,000 consoles dialing all at once. And this is called the reward engine for some of these large games. And so, what happens is, you can create fraud and the job is like, how do you catch fraud in a super dynamic world, where you're literally like, you get rewards as a function of killing monsters in the video game, right?
Dave Vellante
>> Yeah, right.
Alexander Gallego
>> Who can shoot the fastest? And so, what happens is, you can connect machines to these things and then you can sell the weapons for real dollars, like say $10, you get an upgrade to a shield or whatever. And so there are real money laundering use cases here, where you can submit dollars to these engines and then clear it with real dollars on the other end, using this gaming engines as a way to launder money. And so, fraud detection, it's super critical, not just in game to prevent fraud and really to make the game less fun for other users, but also to prevent real world use cases like money laundering. And so, how do you detect that in real time? These people are super sophisticated, the people that are inventing this fraud schemes. And so, that is challenging. There's a bunch of, it's not enough to do geolocation, you have to know their rate. If it looks like a bot, it's probably like a bot, but how do you profile a bot, whatever, across a hundred thousand players, all dialing in all at once.
Dave Vellante
>> And now, is that where AI comes in or is that more a function of your architecture? So, let's talk a little bit about AI. So, that's just a function of Redpanda being, I think, just to recap-
Alexander Gallego
>> Fast.
Dave Vellante
>> We went to market, yeah, superfast, like, NYSE, great example. You can't consume data from NYSE if it's slow, it doesn't make any sense. I think all your traders will leave you. I'll send you your data at midnight, it's not a thing that happens here on the trade floor. But so, what happened with AI, and part of this recent series D announcement and fundraise, is that we acquired a company last year after we talk, that was called Benthos. So, it was a huge connectivity company. And so, what we learned is that, people weren't taking this AI agent to production because they couldn't reliably connect their data sources. So, in order for an agent to do something useful, it has to connect to data. And so, that data has to be mediated, it has to be proxied secured and authenticated and filter. And so, the shift is really twofold. One, people want to understand what these black boxes do, right? These agents. And so, recording that data into Redpanda storage is a key principle. And then, two-Which is an immutable log file, right?
Alexander Gallego
>> Exactly. You nailed it. And the second thing is the connectivity piece, which we bought a company for last year. And so, since then, we really embraced the MCP protocol by Anthropic, and then we added security and filtering. So, yesterday I was at the AI agent conference here in New York. And the main theme there is like, how do I secure access to these agents? How do I know... I don't want this agent to have access to all my Google Drive files. I want it to have access to company kickoff 2025 only. I don't want to have access to whatever, all of the sensitive information that the company has. And so, that was the challenge. And so, sort of the brand extension of Redpanda in this agentic world, is through connectivity, through Bring Your Own Cloud, which we haven't talked about, and through Redpanda being able to be sort of the audit log of all of the events that are happening with these agents.
Dave Vellante
>> Okay. So, that's really your value add, is that you make it fast and you have that audit log that is immutable, Bring Your Own Cloud, including on-prem, I presume, and Edge.
Alexander Gallego
>> Yeah. Yeah. Well, the second thing that we learned is that, okay, this foundation models, OpenAI and Anthropic, they're phenomenal. They're always going to be here. But the problem is, large enterprises, which are our customers, they don't like sharing their data. They're very protective about their data, either for regulatory concerns or privacy or whatever. Maybe it's because we grew up, the people that are in charge of, they just simply don't like to share their data. And so, Bring Your Own Cloud, it's like, okay, let's take this state-of-the-art models, this open-way models like Llama and Mistral and Microsoft Phi-4 and run it inside your cloud. It still deliver as a fully managed cloud, so that's Bring Your Own Cloud. That's an offering. And so, that's it. Those are the three pillars of the company. Bring Your Own Cloud, huge differentiator in the market, the connectivity to 300 connectors, whether it's like Salesforce or NetSuite, and then the storage engine to write this events, the audit, the trails, the observability, the metrics, durably on this kind of scale.
Dave Vellante
>> How do you see this notion of bringing AI to the data, playing out? Because the clouds, they have a lot of AI tools, they have model gardens, et cetera. On-prem has I think been, in many ways, underutilized. And now, because people don't want to share their data, they don't necessarily want to move it to the cloud. Some do, but many are saying, Hey, we're happy with our claim system on-prem, to take an example. What are you seeing in terms of firms wanting to bring AI to the data, on-prem, to build their own AI stacks and how do you fit into that?
Alexander Gallego
>> Yeah, it's twofold. I think on-prem can mean physical, like here in New York or in New Jersey. Those are the older schools-
Dave Vellante
>> Sovereign cloud.
Alexander Gallego
>> Totally. Those are examples. And then there's this hybrid approach where people run their own software still in a hybrid cloud, but they manage the whole life cycle. So, I think it's useful to share that distinction because, people think of, engineers at least, think of both models relatively similar. So, let's take the case where you're here in New Jersey or New York and you have your data centers here connected to the stock exchange. And let's say the use cases, you want to make something smart out of the data feeds that are coming out of the exchange. And so, really what Redpanda gives you is this ability to load the models into the GPUs. We use Ollama as the driving library for it. It's all inside Redpanda Connect, which is the connectivity portfolio, but okay. It's not enough to just execute the model. You have to bring data to the model, and that's where the connectivity comes in. So, those are the two critical ways that we're helping companies that are running inside the data center. And then obviously, is this storage engine, we've powered hundreds and hundreds of large scale customers in a self-hosted fashion. So, I would say those are the three critical aspects that we add.
Dave Vellante
>> So, Kafka replacement, simplifying Kafka is really your real main thrust, is it not? Is that fair?
Alexander Gallego
>> Yeah, that's how we started. I mean, I think that, with the rise of AI and MCP and actually securing connectivity, it's another-
Dave Vellante
>> It's extended....
Alexander Gallego
>> big expansion of the market for Redpanda. But look, the messaging market is massive. It's 30 billion just with Kafka alone and a hundred billion if you take all of the messaging frameworks and products in the market. And so, it's massive and there really aren't that many great competitors out there. And so, we still have, it's a huge market and we're still excited to continue to grow that business. We're just sort of extending the product to AI, to MCP, to connectivity, to help people basically govern agent access to data.
Dave Vellante
>> So, I want to understand the competitive advantage that you bring. I mean Confluent will bundle in Flink, Databricks' Delta Live Tables, so there's competition out there. How do you differentiate? What is your unique value add?
Alexander Gallego
>> Yeah, so I would say, let's start with the storage engine first. So first, we are simply the fastest storage engine that exists in the world for real mission-critical use cases. So, you cannot break a world record today in sports if Redpanda isn't live and operational. I'm talking about 800 microseconds-
Dave Vellante
>> So, cheap fast object buckets, is you're the best?
Alexander Gallego
>> For mission-critical systems, I think you'd be hard-pressed to find something that is better in the market, full stop. And that market is massive, by the way. It's not like it's easy, right?
Dave Vellante
>> No, storage's exploding. I mean, it's just like...
Alexander Gallego
>> Yeah. And so, just keeping up with that complexity side.
Dave Vellante
>> And that ties into open table formats and everything else.
Alexander Gallego
>> You nailed it. You nailed it. And then on the cloud side is, Bring Your Own Cloud. That's another super strong pillar and differentiator. And so, we power the largest healthcare companies in the US, providing technology services for large insurance companies. And so, beyond HIPAA compliance, they just simply... We are HIPAA-compliant cloud, they just don't want to share their data. And so, we're bringing an entire infrastructure inside their cloud and give them a fully managed cloud. So, they love that. And I think, in a BYOC setting, Redpanda stands out as the better BYOC implementation in the world today. And then from the connectivity portfolio, our connectivity portfolio is more than twice the size of any competitor in the market. So, the next best competitor has 150, we have 310 going through 400, later this year, maybe even 500 later this year. So, a huge investment for us. And so those are real hard differentiator pillars for Redpanda.
Dave Vellante
>> And so, the connector hard work you guys did, really has accelerated your business. It's your flywheel, it's differentiated, and it's allowed you to appeal to a much bigger TAM.
Alexander Gallego
>> Yeah, exactly. Because the is data integration. It's like, okay, we sell you Redpanda and people start getting a bunch of value and they're happy with it. But what we saw was that, they weren't leveraging all of the capabilities until we allowed them to bring all their data in to the platform and in many cases, send their data to partners. So, if you're like an ad tech and you want to send their data to Facebook or you want to send their data to whatever, you're like, we have a connector for that. Or if you want to integrate with Salesforce for business process, we probably have a connector for that and so on. And so, really, data integration has been a super strong pillar for us.
Dave Vellante
>> How did you start? You were born in Columbia, you came here in your teens.
Alexander Gallego
>> Yeah.
Dave Vellante
>> How did you start the company?
Alexander Gallego
>> I always identify myself as a builder. And so, growing up, I used to fix dirt bike motorcycles with my uncle at a shop, and those were my formative years. So, I think, my self-identity is that of a builder. And so when I went to school, I was a hacker. We were going to school for cryptography and security and actually hacking of systems and learning how to do that. And then, I had one of the founders of the internet who changed how I think about the world. He's like, "Hey, you're also really good at building. You should build the world instead of breaking it." I was like, that's a great idea. And so, before I started Redpanda, I sold the company to Akamai. I built a really large fast-scale compute framework. So, think of Flink, but in C++, low latency, sounds familiar? And then I left and I wrote Redpanda, frankly for me and it turns out that it got deployed by the largest electric car company, largest CDN, largest ISP in Europe, largest telephone company in Europe. And I was like, well, this probably looks like a business. So, here we are, six years later. It all feels like an accident, but it wasn't. Obviously I built the software that I wish existed and it's been a very organic growth for us.
Dave Vellante
>> Amazing. Congratulations. All right, 12 months out, what are the technical breakthroughs, the market breakthroughs that you're looking for to really help you guys hit escape velocity?
Alexander Gallego
>> So, we are really leaning into this new era of AI agents, really center around securing MCP and connectivity to data sources and things. It turns out, that by default, this older school systems, so things like databases, queues, file systems, and really even cloud services like Google Drive and Salesforce, they weren't designed, the permissioning system wasn't designed for the AI agents era, where you need ultra fine-grained permissions to access. So, the example I like to give is that, I'm user zero of my Google Drive file, right? Because I started the company. And so, we now have hundreds of people. And so, when I connect an agent to my Google Drive, I was like, I'm pretty scared, I don't want them to have access to all those files. I just want them to have access to specific set of files in it, but I want to empower my CISO to inject a policy in this agent that says,, it doesn't matter what the agent ask to the data, there's a boundary layer. And so looking forward over the next 12 years, I hope that Redpanda is seen as a leader in securing data access for agents with MCP, with our suite of connectors. So, expect a bunch of new tools and product launches around just securing agents.
Dave Vellante
>> You've mentioned MCP several times. It's a model context protocol. It's basically a way for agents to connect the tools.
Alexander Gallego
>> Exactly.
Dave Vellante
>> And it's a standard, open standard that Anthropic has put forth. There are competing standards. I think Google has announced one as well.
Alexander Gallego
>> Yeah, A2A.
Dave Vellante
>> A2A, right? And so, we need this in order for agents to have a common sort of language and protocol. Are you betting on MCP or are you sort of the Switzerland and depends on watching the market? We always see these, sort of standards wars emerge and then a de facto wins. How do you see that playing out?
Alexander Gallego
>> Yeah, so, I think it's useful to divorce marketing from what's real in the market. And in my opinion, I think you need one, you don't need both. You can implement A2A as a tool callback in MCP, we've done it. Right, so-
Dave Vellante
>> TCP/IP, that's all we need.
Alexander Gallego
>> Exactly.
Dave Vellante
>> Whether it's the best or not, it's just the standard, run the play.
Alexander Gallego
>> You just need one, right? But I think Google is seeing themselves as a leader, obviously, and which they are. And so, really I think it's more of a competition of who owns the protocol there from a technical level. Now, the bet for us Redpanda is actually a really large pool of connectors. We probably have the best portfolio on data-intensive connectors, and we're adding a new portfolio of what I call less data-intensive. So, think Salesforce, NetSuite, G Drive, Box, those are what we're going out to build this year. But whatever the front-end protocol is, whether it's MCP or anything else, it doesn't matter. So, we build technically a new product that wraps all of our connectors. So, we'll still continue and super-focus on adding more and more connectors as a effort for the company. And look, if A2A wins, then we'll implement it. Right now, we've bet the company on MCP.
Dave Vellante
>> Well, MCP has the momentum. I mean more people are talking about MCP and running, leveraging that protocol, but you never know. But Alex, thanks so much for coming and doing another sit-down with us. Really appreciate it and good luck.
Alexander Gallego
>> Thanks, Dave, for having me. Great to see you.
Dave Vellante
>> My pleasure. All right. And thank you for watching. This is Dave Vellante from the New York Stock Exchange, theCUBE and NYSE Wired studios. We'll be right back, right after this short break.
Dave Vellante