On this episode of the Six Five – On the Road, hosts Daniel Newman and Patrick Moorhead welcome Abhas Ricky, Chief Strategy Officer at Cloudera for a conversation on Cloudera’s enterprise AI ecosystem at Cloudera Elevate 25.
Their discussion covers:
- How Cloudera is helping customers use AI in production
- Cloudera’s strategy to use hybrid to drive growth
- The latest update to Cloudera’s enterprise AI ecosystem
Be sure to subscribe to The Six Five Webcast, so you never miss an episode.
Watch the video here:
Or Listen to the full audio here:
Disclaimer: The Six Five Webcast is for information and entertainment purposes only. Over the course of this webcast, we may talk about companies that are publicly traded and we may even reference that fact and their equity share price, but please do not take anything that we say as a recommendation about what you should do with your investment dollars. We are not investment advisors and we ask that you do not treat us as such.
Transcript:
Patrick Moorhead: The Six Five is on the road here in Miami at Cloudera Sales Kickoff. Dan, the excitement is intoxicating. Everybody seems excited about the new products, the customers, some customers we can’t even talk about that we saw that were super impressive, but it totally makes sense. The intersection of data and AI, we’re here.
Daniel Newman: Yeah, it’s good to be here, Pat. It’s Elevate 2025. You always got to explain the fiscal year to people, but it’s fiscal year 25 for Cloudera and it’s all about the liftoff. It’s all about getting that acceleration and that coiled spring comes from getting product right. But you and I know as entrepreneurs, you and I know as people that are passionate about taking products and getting market fit and getting them to market. Sales makes everything easy for every company on the planet, meaning, there’s never a CEO or chief strategy officer that says, we want less sales.
Patrick Moorhead: Yeah. In fact, I think sometimes you and I even joke that increased sales just makes everything go smoothly. But listen, to completely architect incredible corporate strategy and product strategy, it takes a lot of strategy work. And we just happen to have the Chief Strategy Officer for Cloudera, Abhas, great to have you on the show again.
Abhas Ricky: Yeah, thank you.
Patrick Moorhead: How are you guys?
Abhas Ricky: Very good. Fort Lauderdale as always it’s brilliant in terms of weather. You look great, Pat. As I was telling you earlier, you need to share some of your secrets. And Dan, you’re fancy as always. I wasn’t ignoring you.
Daniel Newman: Yeah, it’s great hair day for me. But no, we do appreciate it and thanks for joining us again. It’s been several times now and we’ve enjoyed really following the journey. There’s a lot of change here at Cloudera, and so we had that slide that said, the more things change, the more things stay the same. Well, you are one of the same in a case where a lot of change and a lot of things are not the same. But overall very encouraging. Spent a lot of time with Charles Sansbury, your new CEO, got to know your new chief product officer well, your new chief marketing officer well.
Great to get to know the team a bit better, seeing some of the energy here at sales kickoff. But, Abhas, one of the things that you actually going back to the previous leadership had worked with us at The Six Five and at our respective analyst firms with the AI strategy. You were kind of trying in an era when Cloudera wasn’t necessarily getting it right, you were kind of the first to say, “I want to mark our territory.” Talk a little bit about how that’s going. Where are you at? How are you accelerating AI and really bringing it to market for Cloudera?
Abhas Ricky: Yeah, absolutely. So there are three angles to it. So one is the customer. So naturally the landscape is evolving faster than we all can imagine. You and I were there at Davos, and if you remember last year in the Promenade, there were two companies were talking about AI. This year, every company is an AI company, whether they sell infrastructure or data sensor or cables, whatever the case might be. So naturally there are a lot of boardroom discussions that are happening. But also the biggest thing is for companies in our space, infrastructure companies, the natural economic buyer has been the CIO or CDO. But in case of AI, oftentimes the discussions go much higher in the organizations, which allows you to become a strategic partner of choice, rather than just a tactical vendor.
And a lot of our companies who’ve been working with us for a long time have actually taken the bet on us, and we have a lot of customers who are running use cases in production, whether that be Porsche, who are running their next-gen electric vehicle design, and after-sales support for the Taycan on our systems or whether that be ABVI or even customers like IQVIA, who were here on stage and who mentioned today how they’re leveraging our platform for the purposes of running large language models, but even doing data governance across 110 countries, and that is a big massive task. So there’s this custom angle, which is driving the pull. The second is in terms of product strategies. The last time we met, we obviously had just announced the enterprise AI ecosystem, and that is still a key position for us, meaning we do want to make sure that we are selling with the ecosystem. AI is not a one-person thing or one vendor does it all.
And there is an enterprise AI foundational stack that I’ve shared before, but largely if you were to ask me to share that of what that looks like. So if you try and follow me at the bottom layer, you’re talking about the compute. So whether that’s the NVIDIAs or the AMDs or the Intel to the world, all the hyperscalers, so public cloud, private cloud, GPUs, CPUs. Then you have the lakehouse, which is where we come in. A lot of our customers tell us that it’s still 80 to 85% of AI is a data engineering problem. So you still have to do the ingestion, you still have to do the data wrangling, you still have to do the metadata governance, and we do that, but then you also want to do fine-tuning and serving and LLM modeling and PEFT and all of that on top of that, which is what Cloudera machine learning is now doing, and naturally we made a lot of progress.
But on top of that, there are specialized capabilities. So for us as Cloudera, we want to leverage the ecosystem there. So for example, we’re working with Pinecone for all things vector databases, vectorization for semantic search is a key capability, and we said we will go forward with that. We announced a partnership with AWS Bedrock, and that is going really strong. A lot of our customers are actually leveraging close-source models from coherent Anthropic Claude or anyone else through a single click in Cloudera machine learning. But at the same time, we build pre-built ready-flow connectors, which allows you to leverage models from Hugging Face. For example, there is a Telco who are running CodeWhisperer and the use case is contact center insights. So they’re collecting the catalogs from the support centers, summarizing them and going forward with that.
So to summarize the foundation stack piece, you have the compute layers, you have the lakehouse, you have all these capabilities, but then on top of that is AI applications. So our strategy in one line is to be able to enable AI applications as much as we can, and that is a key part on the last bit is open source. So the community in general has grown much faster. There are multiple derivatives of Llama 2. There is a scientist from one of the companies at Google in any scale who wrote about the fact that Llama 2, 70 billion parameters one is almost as efficient as GPT-4 for tech summarization, but also it’s 30 times cheaper to run. And I think that is a big thing, which is for large enterprises, you want to run specialized tasks. It’s not just general purpose models.
So yes, GPT-4 and GPT-3.5 Turbo are great for general purpose, but there are other models like Mistral. Now you have some of the Llama derivatives, the long form that has come out as well. They’re very good in terms of doing the specialized task. And as an open source company, we continue to pledge our primary strategy around open source model and the open source community. And we believe that that’s one of the key drivers for us to take that forward as well.
Patrick Moorhead: Yeah, I had never made the connection between your open source heritage and your open source AI strategy, but it makes sense. I mean, that’s what enterprises want, and it is a multivariate world. And talking multivariate world is in terms of where workloads and the data actually sits, still to this day, Cloudera is the only hybrid multicloud solution for data management. You have the cloud only folks, you have some people on-prem only, but you’re still the only game in town, which I think is really important. Can you talk about maybe some recent ways you’re using hybrid to drive growth? I mean, you have this bidirectional, I mean, heck, you’re even partnering with Snowflake in one instance. It’s exciting.
Abhas Ricky: Yeah, absolutely. And it is a great point because nobody in the industry has defined hybrid. I mean, we’ve always said that hybrid is the ability to move data workloads and users bidirectionally without application refactoring costs. And that is important because most customers like HSBC and others will tell us, then it costs between 3 to 5 million per application for rewrites, and it’s a business continuity risk. So we have made strides in terms of replication manager, but also the data mesh strategy that we have that allows you to do that. But the bigger thing is, as you mentioned, we have a very large oil and gas major organization. They’re pulling data sources from on-premises, but they’re putting it on CDP public cloud, and in certain use cases, they’re actually putting it on Snowflake as well.
Patrick Moorhead: Interesting.
Abhas Ricky: So yes, we are working with some partners, but it’s a different flavor of hybrid, because for the customer, the data sources are on-premises, but they’re running the compute on CDP in public cloud. And we’re okay with that because that has allowed us to penetrate net new logos for organizations who were born in the cloud. Because back in the day, if you were one in the cloud, Cloudera wasn’t necessarily a big alternative because our hybrid was on-premises and public cloud, which is still true today for companies who want it. So there are different flavors of hybrid. And then the second point around that is hybrid AI. So IBM talks about hybrid AI in a certain form factor, but we have multiple customers who are actually running large language models in a similar manner that I said, whereby they’re pulling the data sources from one system, but that allows them flexibility and control and they can get away from accidental architectures to be able to define a cloud-native architecture, even for enterprise AI use cases.
Daniel Newman: So Abhas, believe it was in New York, at Evolve. You talked about this enterprise AI ecosystem, and we got to sit down with you, I believe AWS maybe joined us.
Abhas Ricky: Mona from AWS, yes.
Daniel Newman: Yeah, that was a lot of fun. And this was very early days. So in your strategy role, these were the things you’re thinking about, right? I heard you mentioned IBM, you’re pulling these threads, getting the customers to come forward, getting the partners to come forward because there isn’t a single vendor, no matter how much these vendors want to be the only one. You need a combination of tools and enterprise apps and software and hardware and infrastructure and chips. I like to say silicon or semiconductors-
Patrick Moorhead: Chips and SaaS.
Daniel Newman: … eat the world. But can you share a little bit, how is this progressing in the last, was it, it’s been three or four months.
Abhas Ricky: Three or four months. Yeah.
Daniel Newman: How’s that going?
Abhas Ricky: Yeah, I think it’s pretty good. So we made a concerted decision of forming the enterprise ecosystem with a core set of players rather than having a long list of companies who could just come in and participate in this, it’s less of an alliance. It’s more of a strategic partnership whereby we want to integrate with them. So the AWS Bedrock partnership is very strong. As I mentioned, there are a lot of customers who are coming to us and saying, we want to leverage the AWS models on Cloudera machine learning. I actually wrote a blog with Matt Wood, their head of AI explaining that how can you actually go about that. But NVIDIA is another one. So NVIDIA was part of that, we are still continuing to work with them in various capacities, whether that be Spark 3 being available in private cloud in data services, but most importantly hardware acceleration.
Daniel Newman: Right.
Abhas Ricky: And I do want to share that we are coming out with an AI starter kit and that the idea is, and sometime in Q1, and to the extent that I can share the idea is we want to give the customer something to get started and draw value rather quickly. So it’ll consist of four core pieces. They’ll have elements around compute, it’ll have elements around our professional services teams who can get you started and get delivery across use cases such as tech summarization or chat Q&A or Copilot. It’ll have an entire training and enablement module with on-demand videos, reference architectures and all that. But then obviously it’ll also have the product, Cloudera machine learning, and I do believe our hardware acceleration partners with NVIDIA will only help for that.
Pinecone actually released a video on YouTube about how they’re working with us and how semantic search querying is a capability that’s actually helping out a lot of customers as well. So we are continuing to do that. But the interesting thing is we’re also working with some of the other partners. I had a meeting with Google Vertex AI and they’ve shown interest about that. And we do believe that over a period of time we will get to a point whereby some of the other strategic partners will join the ecosystem and hopefully we’ll all collectively be able to help the joint customers we have.
Daniel Newman: Sounds like you made a lot of progress in a few months there. That’s impressive.
Abhas Ricky: Yeah. It’s been a lot of work from the product and R&D organizations, but I also think that, as I said, the natural pull that we’re seeing for customers, and once we start to see the large enterprises deliver production use cases and talk about it, that is driving a lot of the demand. A lot of the inbound is super helpful around that.
Daniel Newman: Well, Abhas, we appreciate you spending a little time with us here at SKO 2025 and actually it’s SKO 2024, Elevate 25, north of Miami, south of Fort Lauderdale in beautiful Hollywood, Florida, on the beach side. But really congratulations on the progress you’ve made. Now, as analysts will tell you, you got a road ahead, you have to prove that you can compete with those born on cloud players. But as Pat and I said in front of the entire sales team today on the stage this morning, you’ve got a lot of provenance. You’ve got a lot of experience and a lot of history, and I look forward to seeing how someone like you, a very strategic thinker, can take that experience, that really impressive customer list and build the innovation and the partnerships to drive it into the future. I want to thank you for sharing a little bit of that here, and I’m sure you’re going to be back with us soon.
Abhas Ricky: No, thank you very much. As always, I love talking to you guys. Always get great insights on all things, technology, but also watches. So I’ll continue to talk to you about that. Dan, you’re Twitter famous about your watches these days, so next time, hopefully we’ll talk about that.
Daniel Newman: That was not me. Thanks, Abhas. And thank you everybody for tuning in. We really appreciate you joining The Six Five here in, not Miami, but Hollywood, Florida. We have had a great day. Hit subscribe. Join us for all the other videos. Check out all the past videos with Abhas here. We’ve had Cloudera a number of times. We look forward to having and chronicling their journey in the future. But for this episode, it’s time to say goodbye. See y’all later.
Author Information
Daniel is the CEO of The Futurum Group. Living his life at the intersection of people and technology, Daniel works with the world’s largest technology brands exploring Digital Transformation and how it is influencing the enterprise.
From the leading edge of AI to global technology policy, Daniel makes the connections between business, people and tech that are required for companies to benefit most from their technology investments. Daniel is a top 5 globally ranked industry analyst and his ideas are regularly cited or shared in television appearances by CNBC, Bloomberg, Wall Street Journal and hundreds of other sites around the world.
A 7x Best-Selling Author including his most recent book “Human/Machine.” Daniel is also a Forbes and MarketWatch (Dow Jones) contributor.
An MBA and Former Graduate Adjunct Faculty, Daniel is an Austin Texas transplant after 40 years in Chicago. His speaking takes him around the world each year as he shares his vision of the role technology will play in our future.
