Powering B2B SaaS Growth: Insights and Innovations

Yeah, happy to join. Thanks for having me.

Yeah, there’s a lot that excites me about it. I love being at early-stage startups. I think when you’re in a company, a Series A company, you’re brought it in as the first marketing hire or brought in to build-out marketing. It’s really greenfield There’s a lot of opportunity to try new things, to be innovative, to build a lot of firsts and build a lot of success in those first things you do, a lot of experimentation. It’s super fun environment to be in if you’re up for the challenge, if you like that dynamic environment. But I’ve been in the startup world now for over 10 years. I would say I was very fortunate to actually work at Databricks prior to Predibase . Databricks is an amazing company on the forefront of data and AI. That really gave me, I I think the bug for this love of startups because I joined there when there was probably a couple of hundred people and got to see truly explosive growth. So by the time I left five years later, the company was about 5,000 people. And I got to see what a startup looks like at each stage of its life cycle from those early days where you’re just, like I described before, what drew me to print that scrappiness of being on small teams, working with your founders hand in hand, working with engineers, a lot of experimentation, wearing a lot of different hats.

When I joined Databricks, I was brought in for product marketing, but one day I might be doing customer marketing and working on a customer keynote for a meetup. The other day, next day, I might be actually planning the execution of that meetup. The day after that, I’ll be working on a launch. So there’s really excitement, I think, early on to wear those hats. And then I got to see Databricks through that growth of early stage to mid-late stage growth. Whereas your company grows, you really start to focus on optimization. It goes from rapid fire experimentation, actually now digging into how can we define our strategy, what is the targeting? It works best? We massage the messaging based on a persona or a vertical level. What are the tactics that really are working best and digging down in there? Got to see that journey like scrappiness of early stage to focus on optimization or later stage startup, and I really enjoyed that. But where my heart is, as I mentioned before, it was really early stage startups. That’s what drew me then from Dane Ricks back into Predibase, where now I got to come in and build marketing from the ground up.

Yeah, I think there’s a few things you need to consider. First, maybe two things early on. If you’re coming in as the first marketer of a high-growth startup or one of the early marketers, the first thing you need to do is really understand the product. What is the product that we’re selling? What is the value of it? And you do that through conversations with the early customers. You need to understand why they’re using the product, the value they get for it. As a marketer, you’re out there telling stories, trying to convince people to use what you have, whatever it is you’re building. So the first thing you do in early stage startup, you got to know those customers. There’s a quote, you should know what school does your early customer’s son go to as an example. You don’t have to be that obsessive. But the point is you really need to dig into why they’re using it, what matters, what’s important to them, understand those personas. That’s job number one. Job number two is like an early marketer, I’d say, in an early stage startup is building the infrastructure so it can actually market. You have the tooling and capabilities to build a funnel, bring leads in, track those leads, automate emails, set up campaigns and webinars landing pages.

Needs some marketing, automation, tooling, maybe some SEO tooling things to track how you’re doing with your content. But you need that infrastructure to go market. Once you have that base knowledge and have the tools needed to be successful, Then I would say there’s maybe two jobs in my mind at early stage company. One of those is breaking through the noise. Predibase, we’re in the generative AI space. This is an incredibly noisy space, super hot space. Everyone’s excited about generative AI, but breaking through that noise, you’ve got to really work hard at that. The key to doing that is building hard-hitting content, knowing your audiences, building content that really resonates with them. We market to developers. With developers, you’ve got to have really things that are rooted in truth and fact and benchmarks, and so showing them like tutorials and ways to do things that helps cut through. I’d say one is filling that content engine so you cut through the noise. The second thing that I think is critically important at an early stage company is really giving your sales teams at bats. When What I mean by that, look, early on, you’ve got to build…

You’re trying to close your first 10, 20 customers. You got to be bringing people in or generating leads so your sales team, qualified leads, your sales teams can have meaningful meetings that hopefully can burn a pipeline and opportunities. For me, it’s once you get the base knowledge and the infrastructure in place, it’s like how are we cutting through the noise with our content and programs, and then how we just have a constant funnel of leads so our sales team isn’t getting bored.

Yeah, for us, at the stage we’re at, we’ve built over the last… I’ve been here about two years. We’ve built a pretty rocking inbound motion. So we have a few things that have contributed to that. So again, as I mentioned before, I’m a very demand-centric marketer. I’m all about creating at-bats and opportunities for my sales team. So we run a webinar every 2-3 weeks. We try to be very educational in nature with that content. Look, we’re marketing developers, right? We can’t just tell them this is cool technology. We have to show them it’s cool technology. We have to show them why it’s cool. We got to show them how it makes a difference. We’re running webinars every 2-3 weeks with either some how-to or showing some new technical capabilities we’ve built. So we do that every two, three weeks, we run the occasional meetup. We had one a few weeks ago. I had a few hundred people in San Francisco. So we have a pretty healthy funnel of just inbound leads coming in. We have a free trial we launched about a year and a half ago. So that combined, every given month, we have anywhere near 500, 1,000 plus leads coming in.

So that’s always going to be a core for me, especially growth stage. You have to keep that and finding ways to grow that efficiently. So that’s always a key part of the strategy, bringing leads in, growing it efficiently, make sure those leads are qualified. I’d say where we’re starting to evolve or think or focus a bit more right now is all on the quality. So we know how to bring leads in at a pretty healthy rate, but now it’s making sure that we’re bringing the right leads in and qualifying them properly. As I mentioned before, we’re in generative AI, so this is a very hot industry. I make a joke all the time that I could throw a webinar down the street in San Francisco about an LLM or generative AI topic, and people are going to sign up. People want to know about this cutting edge technology. So it’s not necessarily always hard to generate inbound or people to come to those programs, but how are we getting the right people to come to those programs? And so there’s things that we’re testing different messaging, we’re testing or targeting the channels we use to market those programs.

And then on the back end is making sure that when the leads come in, we have good scoring our company, so that we’re not just passing everything over our sales team. We’re passing the stuff that is really good because their time is limited, too. They’ve got to focus their time. I want to give them the best leads to work so that they have a higher success rate. For us right now, it’s just now that we have this inbound motion, just, again, trying to focus a bit more to make sure we get the most qualified leads into the pipeline.

Yeah, that’s a great question. It’s interesting because I didn’t really note on it earlier, but when I was at Databricks, early on, I wore many hats. What I ended up focusing on the later years as we got in that later stage growth was actually building out our industry product marketing function. What that was really our charter was, how can we get deeper within these industries, accelerate growth with those industry audiences. What I was doing there was building a team that really focused on how do we take this really cool AI product and all these features that’s built into the product and map that to the use cases and problems that are specific to that industry and talk in the language of that industry. If you’re talking in life sciences, you might be talking to a biostatistician or someone in bioinformatics. They might have slightly different titles, and so you might have to target differently in the language they use. You’re talking to someone who’s doing analysis around some genomics data set, the language they’re going to use is differently. Or if you’re talking in healthcare, the value they’re looking at is maybe patient outcomes as metrics.

The way you talk in your marketing is very critical. Then aligning those capabilities to the use cases really matter quite a bit. You can see what ends up happening when you do that, is you’ll see an increase on your deal size down funnel. You’ll see faster opportunities move faster for the funnel. We start to connect that industry messaging and value to your product. But what I would say is for where we are today, threat of base. We’re still a little bit early for that. To me, that is a bit more of a mid-stage growth. A goal and challenge is how to connect to that next level industry messaging. Where we’re at today is still at that phase of perfectly honing in on our ICP. Who’s our ideal personas? What is the three to four key messages that really resonate with them? We’re seeing good signs in different pockets, but I think when you’re a Series A, you’re always trying to triangulate to find that perfect messaging and persona fits so that you see that crazy lift off. And that’s where we’re at. We’re seeing that growth, but we need to double down on figuring that out.

Once you get that figured out, then you can start to do really cool stuff like, Okay, how do we optimize further for an industry vertical? Or how do we take… Maybe there’s three core personas we’re talking to. How do we build messaging strategies for those different personas and those different industries? I think that’s a bit later for us. I will say where industry messaging maybe comes into play for us today is we do run things like look a like campaigns. We look at our early customers. Let’s just take one. We have a great customer reference, and they’re like a fintech company, some digital bank. If we know their use case and why they’re using our platform, we can use that with our sales team to build them really good outbounding messaging, come up with a really good targeted account list, and even maybe make them a lightweight asset, like a solution brief that they can go market. We’re starting to do more look a like campaigns with some of our initial customers, taking what we learn from those customers and using it to attract similar-looking customers. But that’s probably the depth of where we’ll go with industry messaging at the moment.

You can go way deeper into programs, really integrated campaigns around industries. That’s a bit further from where we are today.

Yeah. Look, SEO, I would say it’s not like whether or not it’s important. It absolutely is important. It should always be top of mind in what you’re doing. I think it’s more about the level of investment in time and money and how you can rally around it. For us, we’re still a pretty small marketing team. There’s three people on the team. It’s myself, product marketer, and a community of developer marketing person. We’re running a lot of… We’re running webinars every two, three weeks. We’re running meetups. We’re publishing a blog every week. We’re building on our website. We’re doing a ton, right? Work with the sales team of our outbounding campaign. So the challenge that I actually face with SEO right now is there’s a lot of great tools like the tools you guys offer out there. We have experimented with our leaders in the past. I really believe in those tools. It’s incredibly helpful when you can understand how you’re performing relative to your competitors, where those gaps are, and then go attack it with your content. The challenge we have is making sure we have enough of that people power in hours to dig in and actually process and parse those insights and turn them into useful information.

Also, I don’t have the deep expertise in some of those tools, admittedly. If you can find someone who’s used them or it can spend a time, I think you get a ton of value out of them. But that aside, it’s still very much part of our thinking of what we do. We use a lot of things like Google Analytics, School Search Console, and some of the other things to, one, First of all, understand, Hey, is any of our content resonating? What are people searching for and finding us? Are those the right things we want to be found for? Sometimes it is. There’s keywords that are very important to us. Where we’re ranking, I’m like, Awesome. There’s other keywords that we built a blog around that turns out is not that important to us. Okay, we don’t need it. We need to diminish those activities, double down on the ones that matter. Then there’s keywords we know we’re not ranking for because we can see it. It’s one that guides our content strategy, just those insights alone, you can get pretty far. Then one that’s to figure out, Where are we? We’re, Why aren’t we building content?

I think especially when you’re marketing developers, content is incredibly critical. Content is key. That is how you break the noise. Are you publishing helpful, hard-hitting content on a regular basis? We’re always looking at what do we want to build content on thematically that is important to us, important to customers, and can help in search. The other one is when you start to see things that work with some of these tools. We had a blog as an example. I know you’ll appreciate this. I’m someone who’s very passionate about SEO. We had a blog where we wrote about one of the most popular models at the time. I think it was one of the maybe meta models, llama models. And we did a tutorial around it. It had just been released. And so there was a lot of noise in the market. People were like, Oh, I want to use this model. And so we did a tutorial, and we made I made sure that from an SEO strategy, the basics were covered. We mentioned the model in the headline of the article or the blog. We made sure it was in our H1, H2 to different places throughout the blog.

We had good density. And that single blog, this was early on in our life cycle, that single blog ended up doubling, tripling our web traffic because it caught wind. It ranked very highly, and it was very focused around a problem people were solving. I want to use this model. I want to use it for this task. We had both those in the headline. And so I think there was a little bit of art and science. We just threw it out there. We knew it was topic was popular. We didn’t was as popular as it was. And then by using those tools I had, I could see it was having an impact on our web traffic. We then re-implement that strategy moving forward. So when a new model came out, we’d go issue a new blog following that same structure, and it continues to deliver. So I think it’s important that you always have SEO top of mind. Whether you can use the tools that are full extent or not, you should be using them, some level of them to get insights, and then using that to unlock these things that are to give you disproportionate growth like we saw.

Absolutely. Honestly, we’re always looking at web traffic. We’re always trying to grow. It started me when I was trying to grow your web traffic. Every single month. It doesn’t always happen, no, but you’re always trying to do it. I’m always looking right at what are the blogs and content and pages that are driving disproportionate growth. I’m just looking at web traffic and where it’s coming in. As I mentioned before, we found that cheat code early on with that blog topic that resonated. It did really well from a ranking, organic ranking perspective. We now use it as a trick. We publish blogs that follow that same structure, and it works well for us. I think finding those cheat codes or hacks and looking at the web traffic and the impact. You got to look at the web traffic and search queries, trying to relate that. So that’s super critical. We look at that. Another one has been pretty interesting for us was looking at… We have a free trial we launched about a year and a half, two years ago. And so we have on the free trial a couple of things. So one, people can self-select.

How did you learn about it? So they’ll fill that in on the form. So I use that data. Plus, I’m also looking at what page they visit before they did a free trial. And that’s been another great insight for us because we can see when certain blog topics If honestly, most blogs at best generate one or two free trials are not big difference makers, but you do hit those ones that have disproportionate growth, where somebody said, Wow, 100 free trials came from this blog. Okay, let’s understand why that’s happening. Let’s see if we can replicate that success. So traffic, looking at some of those, both inbound traffic and search queries, but also looking at how it leads into different CTAs, like free trial, has been incredibly useful.

Honestly, this comes back to the size and the size of your company and to fully leverage these tools. We’re pretty small. And so I try not to do too much investment in tools where we won’t be able to fully utilize them. So it tends to be more the free stuff like Google Search Analytics, Google Search Console, Bing has their own. We have HubSpot under the hood. So we do pipe in data from… There’s some interesting insights to get parsed in there. Like I mentioned, we can see paid traffic to a landing page, a free trial. We’ll look at those things. We do run paid ads, so I’m looking at LinkedIn conversion from time to time or Google conversion. On those ads, I understand is there messaging or things that work well. Obviously, looking at things like email conversion and stuff like that into our webinars and campaigns as well to see how those things are converting. A lot of it is in our basic stack, HubSpot, Salesforce, and all that. There’s a lot more… You can get way more sophisticated, though. There’s definitely things I’m also very interested in that we haven’t quite implemented yet that I think is a future state.

There’s all these great intent signal tools where you can see, okay, are people doing certain behaviours, either on my properties or externally? One popular is looking at developers who are they visiting and contributing to GitHub repositories that are relevant to my business? Should I go target those people? That’s super powerful for sales. Those are areas we’re going to be investing in. And then also we have started implementing things like we get Slack notification if someone’s on our pricing page and doing stuff. So we are starting to implement some of those new capabilities that I think can really give you an edge where you can catch people who are somewhere in the funnel but haven’t made it obvious. Those are some new areas for us that I’m excited about.

Yeah, look, this is a hyper competitive space. There’s a gold rush around generative AI. The minute that OpenAI came out with ChatGPT, that really set the world on fire. People are very excited about LLMs, the opportunity. There’s a lot of new startups. There’s a wide combinator startup every single day that’s trying to capture some portion of the market, and so are we, right? I think for us, how do we differentiate? Look, we made a bet early on. Our bet was that you have these large commercial models that are built by companies like OpenAI, the GPT models. They’re very powerful. They’re good models. I’m not going to tell you they’re bad models. They’re great models, but they’re expensive and they’re very big. When it comes down to when you talk to the customers who are using these tools, often times they have a specific problem they’re trying to solve, a specific use case. You can use a very big, oversize model that’s expensive and slow to solve that use case, but you actually can use… You’re better off taking a small open-source model, which there are many. They’re much cheaper. You own the model because it’s open source, and you can customize it for your use case and get better, at least equivalent results like a GPT-4 or even better.

We made this bet early on that, Hey, we think the future of small models that are smaller, cheaper, faster, and actually more effective for your use case. And we’ve really centred our marketing and product around that. And what’s been really interesting is that the market has really started to lean into this. If you track Apple in the the Apple intelligence platform that they launched recently, or they’re launching in their next to the iPhone, the whole genre of AI capabilities, it’s exactly what they’re talking about, too, which is, Hey, we’ve built these little tiny small models that are really good at individual tasks. And That’s really what our these has been, and we’ve marketed around it. I think just holding true to that and just building our capabilities in support of that vision has helped differentiate us. There’s other people who talk about small models or that are talking about LLMs, but I think we’ve really built a strong story around that and just being focused on it. Then I’d say the other things that set us apart is honestly just providing not only incredible product, but just incredible support to our customers. That’s hard to translate.

You don’t translate to value-based messaging, honestly, in the market. But It resonates with some of the big brands that we’ve done. We’ve won recently. Checker just gave a talk for us at a recent event we held. They’re a background check company. It does a million and a half background checks a month. If you listen to his talk, he talks about all the Predibase has unlocked these small models that have really improved our workflows for how we do background checks, but also talks about the team is just there with me in lockstep, true partnership, helping me solve my problems, very quick to respond to our needs. It’s interesting because those value props, you don’t really talk about your marketing, but I think early stage, your founders have to be embedded in those sales conversations, have to be working with customers. You have to do that to set yourself out apart from these bigger companies that have much bigger teams. I think that’s critically important, although, again, it doesn’t get translated to a value prop we talk about.

Yeah, for sure.

I think we think about developers. They’re very big in community, right? We have some open-source projects like Lorax, which we support. We have a Discord channel and regularly engage with our communities there, posting content, answering, making sure you’re Are you timely? Any customer support or engineering type questions. So having teams that are embedded in those communities, the answer to questions, supporting them. I think a big part is also just like I mentioned before, content is king. So are you regularly publishing content? And my thing about content, obviously as a marketer, all content to some degree is self-serving. We’re trying to get people to want to use our product, but that can’t be the primary or obvious goal of the content. The way I approach content is very much, what is the problem that our community is facing? They’re maybe trying to like, Hey, I want to get better accuracy out of my models. I’m trying to solve this customer support use case, or I’m having trouble building infrastructure, serve these models in real-time. What are those problems they’re trying to solve? And then mapping that back to the things we’re building so that when we have content and having the content tell that story.

When I have content, we always start with, Here’s a problem you’re trying to solve. Here’s what we understand about it. Here’s how you can solve it. Now we’re going to show you how to solve it. And so trying to be very educational in nature, I think is critical. Developers see through that thin veneer of fake marketing where if you just go out there in speeds and feed, say we’re the best and the fastest, that’s not really going to work with them. They have proof is in the pudding. They want to actually see how it works. They want to get in the code. We very much focus our community efforts on building content that helps educate people. Sometimes, honestly, we may be educating them on open source technology, on different techniques that you can do with or without our platform. I think that’s part of it, too. It can’t all just be self-serving. Yeah, we try and map our platform, the problems they have, but also we just try and show them how to solve these problems, how to think of them differently with or without us. I think when you think about community marketing, so content is a really critical piece of that, and that content is told in many vehicles.

You have blogs, you have webinars, We do meetups. We do all those things in some combination, and then video content, too.

Man, we’re so early. I think churn is probably critical when you have hundreds of customers. I think it’s always important, right? I think you have probably bigger customer success in churn strategies when you have hundreds of customers. We’re fortunate that our customers are all very manageable and the relationships are very deep. Churn means a lot to us, right? We don’t want to lose a quality customer. And so we just invest in a lot of those relationships. It’s less of a marketing strategy right now, more of just an organizational strategy. Our founders are embedded, our engineers are embedded there. We have Slack channels open with all of our core customers where we’re helping them be successful. We’re bringing them on as design partners to build new capabilities. It’s a very deeply entrenched relationship at a Series A company. As we get bigger, I’m sure we’ll have more marketing strategy on Churn, but right now, it’s more of a full company press on just being as supportive as possible. I will say I think it’s important, though, at early stage, This is more on the sales side, I guess, specifically. It’s just as important as retaining your customers that matter.

It’s also understanding, is this a good opportunity for us? Is this a good customer? Sometimes saying no is the right thing. Sometimes when you’re early on and you’re trying to get your first set of customers, you have a habit of wanting to be like, yes, we want all of them. We’ll do that. We’ll do that thing. We’ll do that thing. We want that customer. But actually, they may not be the right customer for you. Those capabilities they want may not actually support the broader set of customers that are a good fit for you. It’s also important not only to save the customers that matter, but also say no to maybe the customers that aren’t going to be the best fit for you or even that you maybe can’t even build the best solution for them. You got to make sure that you manage that appropriately. Yeah.

Yeah. Look, maybe this is a bit controversial in a way, but we’re a generative AI company. We’re generative AI infrastructure. I think LLMs, Large Change Model, there’s so many great use cases for them. I don’t think they’re a cure all for every use case. I see a lot of people getting excited about generative content as an example. This is where it might be a little controversial. I think you’re taking the wrong strategy if you’re going to a ChatGPT and having them write you a blog and then think you can pass that off. You’re going to get something that lacks that richness or detail that your audiences crave. You’re going to get something back that wax the soul you need to really connect with your audiences. So I think with generative tools or AI ML, it’s using them in the right way. And there’s a lot of people talking about how to use them, and I believe in the wrong way. An example, I just have this tool, write me a blog, go post it. You’re going to kill your audiences that way. They’re not going to connect to their content. I think the right ways or finding ways to optimize your workflow.

So maybe using generative AI to say, Hey, I’ve written this blog. Help me clean up, change the tone of it, get more of a business owner. Help me, how can I simplify this paragraph that I’ve written? I feel like I might be losing readers too worry. I got to phrase it, How can you help me do that? I think things like that are incredibly helpful. Summarization tasks are great, too. My customer service support cases where, Hey, summarise this phone script into the key points as a marketer, a way to use that is if I’m interviewing a customer to do a case study, use it as a AI ML, a tool to summarize that conversation so I get the key points to help me then write this customer case study. I think that is how we want to use these tools. I think there’s a lot of people going out there with these overstatements, over promises what they can do when the best way is actually to find out how to use them in your workflows to be more effective. We’re not trying to automate ourselves out of a job here. That’s one that wouldn’t be effective, nor does anyone really want that.

How do we find ways to use these tools to be more effective? And so things like using generative tools, improve your writing to summarize things, to shorten workflows. The other area I’m really excited about, we talked about a little bit earlier, is all this signal-based alerts and triaging. So using these tools to identify… When I think about our company and our ICP, there’s probably 10,000 people we can target in North America who could be potential great customers. But not all of them are in the right part of their journey to want to evaluate other tools. Not all of them are truly deep within, maybe, that further along experimenting with tools like ours. And so there’s a lot of these really cool signal alert tools that can look at, okay, how are these different people contributing to GitHub or depositors? Where are they posting on Reddit? What are they out there in the public in these communities saying and the tools they’re using to help you? Then I narrow your focus and your sales team to say, Hey, look, of these 10,000 people we can focus on, here’s 500 who we know are doing the activities that are wrestling with the problems that we solve because we see the activities they’re doing online or they raise their hands in different ways that aren’t as obvious as filling out a lead gen form on a website.

They’re doing things online in communities. How can we use that to then target our efforts towards them to help of bringing them into our funnel? Those are tools I’m super excited about. There’s marketers who are doing great work around that. That’s an area that we’re going to be investing more in.

Growth, baby. I think you look at your Series 8 company, it’s all about ICP fitter than growth. So we found some really great traction. We’re seeing good growth. Now, I just want to put the gasoline on the fire and figure out how we just accelerate that even more. So growth is just incredibly important. Continue to grow our funnel, get more qualified leads in, more meetings for sales, more opportunities. I’d say a big focus I mentioned earlier in our conversation is all around quality. So we found some We create strategies that work for us to bring leads in. We’re in an industry where people are trying to learn. So there’s a lot of people kicking tires. We’ll produce content, they’re going to read it. But how do we actually get to the people that are doing the activities that make them well suited for a platform like this? Or somewhere in the buying journey, we’re They’re ready to try us. So really just figuring out how to find those people better, message them better. And then when we get all these leads in our funnel, continuing to really focus on scoring them properly so that we’re giving our sales team the best leads to follow up with and they’re not wasting their time.

Yeah, let’s go. Bring it on. All right.

Oh, man. My favorite marketer. There’s someone recently who I’ve been following that I think is pretty savvy guy, and I’d recommend everyone check him out. His name is Kevin White. He works at a company called Common Room. Common Room is a tool. I actually agree. I’ve only watched your videos. I haven’t actually used it. I’ve spent a lot of time investigating it. But Common Room actually started as a community tool. I think managing community is what it does. It’s all that signal-based. They have 100 plus different signals you can surface to help you target your marketing efforts. Kevin White works there. He is someone who I really see as on the cutting edge of marketing intelligence and the signal-based targeting and all that. He’s always posting just incredibly useful stuff, great insights. So someone that I’ve started following the last year would highly recommend if you want to see what’s on the bleeding edge of some of this new ABM and marketing automation of how you can improve your targeting.

The weirdest thing I have done in the name of marketing. Oh, man, I’m sure there’s something strange. I’m trying to think of all the trade shows we I’ve been at. I don’t know. We’ve probably run some weird contests, beat forth weird prizes. I’m trying to think if we ever gave anything. I’m drawing a blank. I’m sure there’s something weird we’ve done. I’ve never had to put on a mascot suit or anything. Salesforce, Dreamforce, I have. There was no software. People in the mascots with the big red X. I’ve never had to do anything like that. We’ve talked about it. I’ve never had to do it. One of your Databricks, I wasn’t involved, but they had Databricks at our big conference when I was there. They had that Wack-a-Mole game. In the arcades, you wack the They had one of those and it was like, whacking big data problems or something. That was pretty fun. I didn’t do it, but that was cool. I’m sure there’s something.

My Favorite brand mascot? Man, I don’t know. Who’s got a cool mascot? Does anybody have a cool mascot? I don’t know. I can’t think of one, to be honest. I don’t know. I’ve got a good mascot. It’s funny. At Predibase, we’re a very serious brand in a way. We’re very focused on research and technology. I see a lot of these brands, especially in the generative space, who have llamas associated with them, because if you follow LLMs, Meta has one of the most popular LLMs. It’s called llama. You see all these startups where their logo is a llama. I’m not suggesting we do that in front of me, but it does make it fun when you have a cool guy like that. I would say our open-source project, Lorax, has a pretty cool little guy. He’s pretty fun, so that’s cool. So our Lorax mascot is fun. People should check him out, Lorax Change. If you type that in, you’ll see him. That’s a cool one.

I don’t know. Really? Nice.

My what? My last prompt? Yeah. Wait, would you… My last prompt was probably actually like… It was something like writing a social post or a blog. The use case that you were earlier where I was having a model actually help me simplify the language in a blog. How can I communicate this paragraph better. If you follow our LinkedIn page, so one of the things that there’s a new open source model that comes out almost every couple of weeks. It’s right when it comes out every day, but one notable, I’d say, model that comes out every two to four weeks. When those new models come out, like llama 3.1 or Phy, whatever, Microsoft FI, I2, whatever they are. When a new model comes out, people are always super excited to try them out. So one of the first things we do, we have a free trial. We put on a platform so people can go prompt them and try it out. So one of the things I always do, I have a running joke that I typically… I’ll do a gif to show people, Hey, we have this model on our platform. Go try it out and post on social.

People like gifs, right? So I’ll constantly put in there. I’ll prompt the model on our platform. Like, How many llamas does it take to screw in a light bulb or something? It’s a bad dad joke. And the responses are usually like… But I use it consistently for whatever reason. I did it once. I was like, I like that. So if you see our LinkedIn page, next time we have a new model, I’m sure it’ll be a gif where you’ll see our cool platform, how to prompt those models and what you can do with them, and you’ll see that prompt. How many llamas does it take to screw in a light bulb and some really cheesy answer.

That never dies. Jeez, never dies. I have a hard time envisioning some of the things that I’m really passionate about ever dying. I’m very passionate about content. It is how you break through the noise. It is how you deliver value to the world. Man, I guess my hope is that there’s so much noise today in email, on LinkedIn, online. We’re so stimulated with messages, and now with generative content. Actually, people, as I mentioned before, are using these tools incorrectly to generate just solely List content or probably factually incorrect content, not using schools properly, where you should be using the optimizer content as a comparison. I hope we don’t get so awash with bad content. The good stuff can’t rise to the top. I have faith in humanity. Good content will always rise to the top. Reddit has bring that model out that people will upvote the stuff they like, although they often up vote the funniest thing. But good content. I hope good content continues to reign supreme. That’s how we communicate with our audiences and show value. I hope content continues, and I don’t ever see it going away, but there’s a lot of competition to break through.

So I hope we don’t lose good content.

Keep your content human with a touch of AI. Let’s make it better with AI, but let’s keep it human at the core.

Absolutely. It’s been a blast. Thanks for having me. It’s a pleasure.