Managing AI as Workforce: Control, Accountability, and Governance — CEO Roundtable

Julia Nimchinski:
Awesome, thanks. And next up, let’s talk about managing AI agents. Welcome to the show, Mary Shea, co-founder and CGO of Beercat. And we’re gonna have a phenomenal session here. top CXOs of the industry. Mary, take it away.

Mary Shea:
Fantastic! Well, it’s so great to be here. It feels like it’s been too long since we’ve had one of these, Julia, so thanks for bringing us back, and… I just want to say kudos to the two guests right before us. I just came into the last because I’m struggling with some of those issues that were discussed. But anyway, we’re here to really talk about managing AI as a teammate, as part of the workforce, and we have an amazing panel. I’d love to have each of our guests introduce themselves briefly, and then, I’d love for you to talk a little bit, just give us a really. quick elevator pitch about your company, because I think that’s going to help the audience really understand what your perspective is and the point of view, that you’re going to bring to some of the questions that I’m going to put forth. So, let’s start, left to right, and Xander, why don’t you, kick it off for us?

Zander Pease:
Hey everyone, great to be here. I’m Xander, co-founder and CTO of Spara. We’re one platform to build, deploy, and optimize go-to-market AI agents. And perhaps more importantly, I want to talk about our vision for what that means. So we don’t believe that AI is ever going to fully automate sales. It’s one of those last areas where human touch is so important. And so because of that, we’ve built SPAR as a platform that we think gives you the best of both worlds to have a hybrid approach. SPAR can deploy agents that are going to automate buyer-facing touchpoints that your salespeople shouldn’t be focused on, but then the other half of that is AI agents that assist your sales rep both in the work and in the buyer-facing touchpoints that they should be focused on.

Mary Shea:
Awesome. Thank you so much for that. And, let’s go to Amanda. It’s been a while since we’ve run into each other, so it’s great to see you, and I was doing a little deep dive on your company, and like all of the other companies here, it’s absolutely fascinating. I can’t wait to hear from you. So, give us a little bit of a background, and then your blurb, and we’ll get on to Dipinder.

Amanda Kahlow:
Yeah, sounds great. Great to see you again, Mary. Yeah, I’m Amanda Kalo. I am the CEO and founder of OneMind. I was the founder and former CEO of Sixth Sense. My one mind to the market is I started Sixth Sense to find buyers and One Mind to close them. We are building what we call very different from Xander, in the sense that we are building go-to-market superhumans, and we are leaning into… we are replacing roles across go-to-market. everything from inbound, SDR, which is the obvious on the website, to we have sales engineers that join calls, that can give live demos, that can solution, like your best solution seller and sales engineer, all the way through to onboarding, through in PLG companies, and have, like, long tail. She can go all the way, click around on the screen, share the screen, read the screen, do what a human does. And then, you know, for some companies, for our customers, like HubSpot and others, that they, for their SMB and commercial business, she goes all the way to close. So, we are really leaning into the fact that I think the last role to go is the AE, but we are absolutely saying that AI will replace, a lot of roles across go-to-market.

Mary Shea:
Wow, I just kind of want to dig into that right now, but I’m going to put that on hold, because that’s a little bit of a controversial, topic which many of the experts that I speak to on a regular basis shy away from, so I like that you had the balls to just kind of put that out there, and we’re going to come back to that. So, Devinder, it’s great to see you again. Give us a quick background and tell us a little bit about your company.

Deepinder Singh Dhingra:
Absolutely, thanks for that, Mary, and hi everyone, I’m Dipander Dingra, I’m the founder and CEO of Refter.ai. We are a full-funnel agent AI solution for B2B go-to-market teams, especially B2B go-to-market teams with complex GTM motions. We focus on helping upper-mid market and enterprise B2B companies, and our focus is really about how you take the fragmented GTM activity across multiple systems and turn that chaos into predictable revenue growth. So, key aspect of that is how do you deploy coordinated AI agents for every stage of the B2B buyer journey, right from the anonymous visitor, through to the different stages, to close one and beyond. I’m very excited to be a part of the panel, because this is what I think about day in and day out. Yeah.

Mary Shea:
Me too. Thank you for that, Defender. And, Noam, off to you.

Noam Schwartz:
Hi. Thank you so much for having me. My name is Noam Schwartz, I’m CEO and co-founder of Alice.io. Some of us maybe knows us as, in our previous name, Actifense. one of the early vendors in the AI security space, and we’re solving a problem that every company that is racing to deploy AI has right now. Everybody tried to have, like, AI in production in front of customers. To help their business do better. But most don’t really have an idea what the systems are actually doing, what… who they’re talking with. what they’re leaking, and we give the teams, the security teams, and product teams the visibility, the control, and everything they need in order to understand their AI layer. To catch the threats, to catch the misuse by adversarial actors or by mistakes. Understand the data exposure. before they become incidents, before they become PR issues, before they become, like, the next crisis. And think about us as the security platform that is built for this age, as Amanda said, that AI agents are doing so much of what we’re doing today, so enterprises can actually move fast, but without being, you know, flying blind.

Mary Shea:
Perfect. Thanks so much for that, Noam, and it’s great to have you, sort of the primary expert here on security and compliance. It’s something that’s a super, important topic. And Afshan, if you don’t mind giving us a little bit of your background and, your companies.

afshinnikzad:
Sure thing. Thanks for having me. I’m Action CTO at Cross. So, and also co-founder, so in across, the co-founders, built across, based off of their research from Stanford and UC Berkeley, we are Stanford PhDs, UC Berkeley professors, or thesis and across is that, in able to produce value in enterprise, agents need to Convert context? To execution, not just by magic, but by a layer that does that transformation. This goes back to our research, which was focused on reliability in AI, and basically, based on that research, we built this context-to-execution layer with a starting point from GTM teams, processes that are as high a stake as end-to-end sales. need a thesis, a system, for generating actions that are grounded in context and bound by permissions in enterprise. That’s what we have built in across.

Mary Shea:
Awesome. Thank you so much for that, and Let’s move forward. I’m going to just give everybody a couple of thoughts that I have, and then I want to just dive into some of the questions, and I think we’re going to have a really great session here today, but for those of you who actually follow my work, over the last year or so, I’ve been spending a lot of time really thinking about and talking about AI as this productivity layer. You know, tools that help sellers move faster, write better emails, summarize calls, but I think we’re moving beyond that now to where we’re really starting to see. That there’s a structured shift in the way Whoops. Sorry about that. Work gets done, and the AI now is being integrated into Teams as, an actual teammate, and that’s gonna have all kinds of implications on. Rules and responsibilities, who owns the work? How do we define and outline accountability? How do we find the handoffs between the AI and the human sellers? And then, the governance piece, how do you really, actually govern performance at a scalable level? So, I think really what I want to do is start to delve into, in this conversation. AI as a workforce capacity, as a teammate. And so, I really want to start out with you, Amanda, because I’d love to hear, really, how you’re framing, this shift. You’re not talking about co-pilots, you’re talking about teammates, and you’re actually going right out on a limb and saying, you know, we’re going to replace just about everyone, except for the AE, and I certainly have a perspective on this as well, but you’re the expert. I want to hear how you frame this up, you know, for the C-suite when you’re having those conversations.

Amanda Kahlow:
Yeah, I mean, I think we live in a world, as we all know, we’re all being squeezed, right? Growth is more expensive it costs, you know, exponentially $3 to acquire a customer, what used to cost $1.50 for a dollar of ARR. Our growth rates are shrinking, and the board is demanding more. Your customers are demanding better. If you think about the handoff from your website, to an SDR, to an AE, to a solutions engineer, to a sales engineer, it’s atrocious. Like, the process we put our buyers through as humans is really, antiquated. And there’s… humans don’t have good memory, recall, capacity, time limitations, etc. So, I think from a growth perspective, an efficiency perspective, the market and every… we all demand better, and… at the end of the day. And so, we’re looking at this as where can AI come in and accelerate growth and give a better buying experience? And so the obvious, I look at it as two ends of the spectrum, so our… Superhumans are working for, you know, an inbound on the website, but it’s not just acting as, like, a basic chatbot that’s booking meetings. It’s solutioning, it’s giving the pitch, it’s giving the full demo, so it’s acting as a sales engineer in that first moment, on that first touch. driving the conversation forward and shortening sales cycles. And then when it does get to an AE, and I think AE is the last job to go, I actually think the navigating the org and the relationship building, that will be here for a little while longer, but On the other side of the spectrum is your sales engineer or your CSM that’s onboarding customers and taking them through a series of steps and repeatable tasks. And AIs, or what we call superhumans, can do that exponentially better than a human, and it’s always available. You don’t have to schedule. You can, like, think of one of our customers as Alteryx. They have 10,000 case studies in their superhuman’s brain. A human could never have 10,000 case studies ready to go at any moment. You have one deck of, like, 20 slides, hopeful that the right question comes in. You don’t know where the conversation’s gonna go. So the… the capacity limitations of human across these conversations are just really poor, and so AI just can do so much better, and it’s not about automating the tasks, it’s about taking on the full job, and multiple jobs, not just one. So, you know, when we think about on the website, it’s the chatbot, it’s the SDR, it’s the solutions engineer, and it’s the first call AE, all on that website.

Mary Shea:
Yeah, I’m… I mean, I love that. I’m… I’m sort of, dialed in to, a phrase that you said, which is the AE is the last job to go, and I actually was chatting with Xander earlier this morning, and I believe… take this on over a coffee or a glass of wine, but I believe that, you know, in enterprise B2B sales, that it’s a very emotional buy, and that You know, ultimately, you want a human to be accountable, for, the success of your career in the program. However, I’ve seen what’s on your website, and I think it’s very, very exciting. I actually think that the AE role changes dramatically, and it becomes more of, like, a COO, right? So, you’re less like the relationship guy or gal, and you’re more, an operator who’s managing, you know, 20, 30 different inputs, right, and making and instrumenting them and making those decisions. So, that’s taking us a little off track, but, it’s such an important and interesting conversation. So, how are you coaching and talking to leaders about, how they should think about capacity, org structure, augmenting humans? Like, how how is this going to impact the overall go-to-market organizational structure? I mean, that’s kind of a big question, but I’d love you to weigh in, as well as anyone else who might have a point of view on this.

Amanda Kahlow:
Yeah, I mean, I think the org structure is fundamentally shifting. Like, I’ll just think of our own example. We have no SDRs, I have no sales engineers, and I am 18 months into market at the same revenue that I was at 6 months… at Sixth Sense at year 6. Right? So, we do still have… Like, it’s… it’s… it’s wild, because we are, and we have… our superhuman is touching and opening 80% of our opportunities. She’s having that conversation, and she’s supporting the deals in the cycle, right? So now my AEs, like you’re saying, gets to be that control tower, right? The AE is actually managing the relationship, saying, who else do I go to in the org? But she’s not… she or he is not the disseminator of information. She’s not selling, and she’s not solutioning. Our superhumans are doing the solutioning. They’re joining the calls, they’re asking the critical questions, they’re finding the pain and the impact. And they’re doing the things, and then our AEs can help move the deal forward, who else we need to get to in the organization. And so there’s those pieces that the AE’s doing. So I think that still lives right now. But I will challenge you that I don’t think humans want to buy from humans. I actually think that’s a.

Mary Shea:
Valiseave?

Amanda Kahlow:
Yeah, I actually… what I think is happening right now is trust. Right now, humans trust humans more than they trust AI, because if you go to the foundational models and the LLMs, they are just probabilistic models, and they’re hallucinating, that’s their job. But when you put it on tight guardrails, and you have the evals, and you’re doing application-specific AI, it does not hallucinate. So the day that the buyers realize that the AI is less likely to vaccinate than a human. There will be a massive shift, and your buyers will demand to talk to the AI, because they’re going to trust it more than they trust the human. That shift hasn’t happened, but it’s coming.

Mary Shea:
Wow, I love that. I love you putting a stake in the ground like that. You know, I understand the logic. Xander, do you want to weigh in? Do you have a perspective? I know we were kicking this topic back a little bit earlier while we were having our coffee.

Zander Pease:
Yeah, I’m happy to. This is a great topic. So I guess to start, I agree with almost everything Amanda said. I think the difference in our platforms is that while our agents can do the full automation of the sales cycle, we also have this entire assist piece. And the reason that’s so important for the market today, and we think probably forever, is that the go-to-market motion is not going to be 100% Run by AI, depending on your product, depending on your type of lead, your type of conversation. So we’re going into the market with the opinion that if you want to automate the whole thing, great, you can do that on Spara. If you’re not there yet, great, you can do that on Sparra. If you have some buyers where you want to do that, great, but the high-value buyers you have a different motion for. As the market changes over time, as the capabilities of the AI changes over time, you can A-B test and sort of update your motion really, seamlessly, depending on the tech, depending on the market. That’s how we think the world is going to evolve in the near term.

Mary Shea:
That’s interesting, I’m kind of, like, obsessing on this topic now, so I promise I will move on, but Amanda, give me a time frame. What year is it going to be that buyers trust agents more than they trust human sellers?

Amanda Kahlow:
I think it’s this year, 100%. I think we’re already moving.

Mary Shea:
I mean, to that one.

Amanda Kahlow:
world, where they’re starting to expect that they… they’ll… they’ll expect that experience, because when you have that one aha moment, like, if you’re calling, like, I called a plumber the other day, and it was an AI, and it was so delightful. I was like, no, it’s this… and it understood me, and I was like, no, I need it, I’m gonna be there on the weekends, my, like, when I’m in that house, da-da-da, and it was so much better than talking to a human. When we realize that AI can actually do this better than a human, I’m going to be frustrated when I get the put to the human AI, and now I’m talking B2C world, right? To make it relatable. But I think this chasm is shifting, and it’s shifting fast, as, you know, great companies like these are building AI agents that aren’t hallucinating, and actually have more depth. So it’s not even the hallucination, because humans hallucinate freaking all the time. Like, when does seller not hallucinate to get a deal done. Like, I love it when people ask me if my AI’s gonna hallucinate. I’m like, do your salespeople hallucinate? But it’s the depth of knowledge. So, when, you know, you have to build an AI agent for, like, we’re about to work with a very large data company, enterprise data company, and they sell data to every industry and every vertical. They can’t hire the right solutions engineers and sales engineers that can understand, have that depth of knowledge. And so, AI can, and it can solution, and it can bob, and it can weave, and it has memory in a way that a human can’t, no matter how good we are. Even the CEO can’t, right? So, it doesn’t understand every vertical that I sell to. Like, we all have, like, we all have guardrails on what we’re capable of.

Mary Shea:
Yeah, so I guess what Amanda’s saying is that, her agent’s gonna come for her job, too. I just heard that there.

Amanda Kahlow:
The day that it can do my job, I will hand the keys willingly. And I even say to my team, if your job gets replaced, that we will forward vest your equity if you replace yourself. And now, I haven’t approved this with my board yet, but I’m really working on this, where we can forward-vest your equity, and we’re gonna find another role for you, right? So if you want to stay, or you just forward vest and, like, sit on a beach.

Mary Shea:
There you go, there you go. So we’ve got some, questions from the audience, which I want to make sure we answer. Real quickly, you know, are you talking about just one agent or multiple agents, Amanda, as, as we go through, the buy-side process?

Amanda Kahlow:
Right now, I mean, it’s an interesting question, because technically, every superhuman is about 15 agents running, background agents, running behind the scenes. Alright, so I’ve got the one that’s gonna give the demo, I have the one that’s qualifying, I have the one that’s pulling information, and then they’re swapping each other in. It’s not just creating one, and it’s just to keep the context down and make sure we can keep the speed, and the latency is really… and cost is really important in our world. But the way we position this to the market is… each superhuman takes on a job, and it’s… think about it as the content that you’re going to put in the superhuman’s brain. You wouldn’t put your customer, CSM, you know, content into your inbound, you know, agent, right? share that level of detail, or a sales engineer has more, you know, integrations and technical and security information that you don’t just give top of funnel so all your competitors can come grab it. So, as you need to, like, wall this off, we think of those as different superhumans, but each superhuman can have unlimited flows and goals, like. Context-aware, based on past conversations, can go in a different direction from a specific campaign and be hyper-personalized and contextual to who their buyer is.

Mary Shea:
Awesome, thank you for that. I think… I think this is a really good, handoff to Noam. So, we’ve got another audience question, which is, how are the guardrails being implemented so human trust can be achieved? So, yeah, I mean. tell us, like, what changes the moment AI starts operating at that level of autonomy that Amanda’s talking about? What happens inside your organization, and what do you need to be worrying about?

Noam Schwartz:
So I really liked what Amanda is saying, and there’s two things here. First, I totally agree that this is the year. I actually think I think it already happened, we just, like, many of us didn’t really notice. Because when… Effectively, we’re trying to find that trust moment. And the speed of which people Can shift who they trust, what they trust, is where there’s, quality. Where there’s consistency in that quality. when there’s reliability, and when there’s accountability. And, you know, it’s like, I’m right now in San Francisco, there’s RSA, it’s… everybody’s, like, driving, like, self-driving cars in Waymo. And the second, like, the moment that people get into the car, it’s, like, frightening in the first moment, and then it’s kind of… it’s like it was here since forever. It’s… it’s my right to drive in a self-driving car. They trust the car. It took 5 minutes. From… oh my god, it’s sci-fi? to, yeah, I’m… this… this is… this is Wednesday. Like, this, this is a regular day. And, and… we’re also, like, in our company, we’re in an event right now, we have, like, a WhatsApp group that we’re coordinating with everybody on the booth, on the ground, networking event, and in that WhatsApp group, there’s an agent. That helps us coordinate. Like, hey, who’s in the booth shift? Who’s going to that networking event? Who met that customer? Did we… did we log the… the business card? It’s all happening, it’s all, it’s all happening right now, and my team can trust it 100%. I trust it 100%. And now, why do I trust it in 100%? It’s because That we’re making sure that the accountability model won’t break. This is what we’re solving. There’s… there’s… folks that are building the agents are thinking about the quality. They’re thinking about the consistency, they’re thinking about the SLA, they’re making sure that it will do what it’s supposed to do. We’re making sure that these products will also be accountable, and they won’t break. Because the moment they’re broken, first, you lose trust, because it’s a new technology, but also a lot of the naysayers in the company, would kind of, like, use this to say, hey, you saw This thing broke, or we can get into trouble. And it will slow down adoption. And the whole thing about making sure that you can understand what’s the chain of accountability, why the agent is doing what it’s doing, what context, on whose behalf. What’s the actual operation is the most important thing when AI is operating with real autonomy. Making sure the accountability model won’t break.

Mary Shea:
So, so let’s get a little bit more specific here now. I mean, thank you for that, but so, so… What? What are the specific controls that you need to put in place to make sure that, you know. You, you don’t have a disaster within your organization.

Noam Schwartz:
So, first, who’s using the agent? The agent is working on behalf of who? Like, is it associated with a team, organization, an individual? This is very important, so you can even trace back what the model is supposed to do. Is it, like, it’s… who told it, what to do? Then there’s the instructions. How do you define what the model’s supposed to do, and what the model absolutely shouldn’t do? There’s a layer of… Policy in the organization that you bring to any employee that starts. Sometimes it’s part of the onboarding, sometimes it’s part of the training, sometimes it’s just… knowledge that runs in the organization, but how do you translate that into an agent, into an autonomous agent? So that piece. Then you continuously need to harden that, to make sure that it’s protected. Against the most recent, the most common, but sometimes the novel ways that, bad actors are trying to, manipulate the agent. They’re trying to steal data from your organization. They’re trying to sometimes steal money from the organization. So how do you make sure you have that layer? Like a red teaming for your employees, like, like, like, like security training. And basically translating your company’s policy, your company’s red lines, into an automatic agent is not trivial, because we don’t have that documented to most companies. The voice of United Airlines is very different from the voice of Virgin. They sound very different. If you use an agent that is working exactly the same for two companies, they just sound the same. And they don’t have the same policy, they don’t have the same risk awareness, they don’t have the same Q&A, They’re very different, so how do you build this? And guardrails is not a one-size-fits-them-all. You need to have your own guardrails, because you have your own personality, you have your own conscience, you have your own brand story. And this needs to come into place also in your guardrails. It’s your voice, effectively.

Mary Shea:
Got it, got it. Thank you for that. And, let’s move on to Afshan. I’d love to get a little bit more technical, right now, and so, you know. What has to happen from a technical perspective to move to really having AI as a trusted operator within the company, or within the organization, within the go-to-market team?

afshinnikzad:
Yeah, so… You know, models are pretty powerful. If an agent have access to them. For simple actions, like… Summarize a call, or draft me an email, like a response email. Agents don’t have a lot of challenges. But when it comes to executing a process end-to-end, a complex, cross-functional, adaptive process. then the trouble starts. I’ll give you a true story. We were working with a Fortune 500 company. They, before starting to work with us, they built an agent that they tasked it with increasing top of funnel. The agents started canceling existing subscriptions. I guess the argument was that any existing subscription after being canceled can be automatically counted as the top of funnel, right? So it did the right thing, except that it did not have the right context. It did not understand the big picture.

Zander Pease:
So…

afshinnikzad:
In these processes, we take One thing for granted, and that is how context is generated.

Zander Pease:
To act…

afshinnikzad:
When these actions are simple, think about choosing between A and B. Like, could I route this support ticket to this team or that team, right? Maybe that simplistic approach works, but when it comes to more complex actions. that have infinite degree of freedom. Think about drafting a POC proposal. That kind of ad hoc system that let’s put everything inside a prompt and ask for an outcome, that doesn’t cut it in terms of enterprise standards. It doesn’t have consistency. If you run the same process twice, you get different POC proposals. it’s not necessarily explainable. You wouldn’t know how the POC scope is defined, right? So… for us. what unlocked… Building enterprise agents. Was, building reasoning graphs. Which is basically a… Context to execution system. It ingests data through integrations, Reasons over data in a basically layered manner. and then helps the agents to decide. For example, the agent who is generating that POC proposal knows that it needs to understand the POC scope, and it knows that to decide the right scope, it needs to not understand the solutions we can offer to this customer, but also What we… if there is a competing account, if there is a competition in this account, what we can do better than that competition. And based on this multi-layered reasoning approach, can make actions. This makes it also explainable and consistent. So, basically, I think what is underestimated in enterprise deployments now is this layer. How context is there into actions. And I think successful deployments need to work on that.

Mary Shea:
Thank you, thank you for that really thoughtful, thoughtful response. Amanda, let’s go back to, you know, sort of the topic that we were delving into, with, your thread here. So, the AI is really doing the real sales work, the qualifications, the demos, the sales engineer, the interacting and engaging. So, who’s the boss of the AI? Is it… is it the human seller? Is it the sales enablement lead? Is it the CRO? I’m like, who’s… who’s on the freaking hook?

Amanda Kahlow:
I love that. Well, it depends what this… which… what superhuman you deploy, but I’ll give, like, a few specific examples. So, for… for example, we have our own sales engineer superhuman, right? So, I actually have a sales engineer that’s job it is to just fill the brain. He doesn’t join calls with our customers, prospects. His job is to make sure the superhuman is on… on the guardrails, has all the latest information, and you know, like, if we think about the pace of innovation today in this world of AI, one of the hardest things across go-to-market is the relationship between product and sales, and the relationship between product and CS. Keeping up with all the features and everything that the company… like, I can’t even keep up with what we’re doing. We’re releasing so much, so fast. that, you know, keeping that information that comes from product, this is what we built, to put in the superhuman’s brain, that’s the job of our sales engineer that is feeding the brain and making sure that’s consistent and stays on track. Now, from a sales perspective, when you’re talking about, like, closing revenue. and who’s on… who’s in charge of the revenue targets, you know, I think that’s… it has to be top-down, right? So it’s now, instead of looking at things, we’ve holistically looked at things from a marketing, a sales, and a CS perspective. Even though you have a CRO sometimes that goes above all three.

Mary Shea:
Right.

Amanda Kahlow:
CRO is sales, but what I think that’s going to happen in this world is that’s going to collapse. Where there actually is somebody who owns the full go-to-market life cycle, and I think these executives own the life cycle, need to be measuring to the bottom of the funnel. Are we shortening deal cycles? Are we increasing the AB, and are increasing revenue? Not increasing MQLs at the top. And so all of these chatbots, even, like, chatbots that have a face now, still just book meetings with an SDR. Right? And so, it’s… we have to be looking at what is the impact that the board cares about? What’s the impact on our business? And the cross-functional impact? And I think there’s going to be a huge restructuring of organizations, of really, truly having a CRO, and having somebody who owns the whole end-to-end. life cycle, because that’s in the world of where AI can do it all, that’s who’s going to be responsible at the end of the day.

Mary Shea:
Yeah, we’ve had CRO… the CRO role has been in play for, what, 15, 20 years now, right? And back when I was actually CRO at that time, it was sort of a role in name only and theory only, right? This was the vision.

Amanda Kahlow:
Right.

Mary Shea:
we weren’t able to truly execute on that vision, and I think what you’re saying now is that vision is going to come to life. I was just speaking with Seth Morris the other day, and he said the exact same thing, which is, you know, there’s going to be one executive that oversees everything go-to-market, and why wouldn’t… why wouldn’t there be?

Amanda Kahlow:
brain that takes the.

Mary Shea:
Yeah. Right? It’s one break.

Amanda Kahlow:
Think of it like there’s one set of core corpus of knowledge of, like, we call it an account-based context graph that goes across the whole life cycle that just supports different touchpoints and different needs of your buyers. You don’t want to have these things in silos. That’s not how your buyers buy and how they think. They’re still the same buyer with the same needs from this conversation all the way through to upsell and cross-sell.

Mary Shea:
Yeah, exactly. So what’s your, what’s your time horizon on, sort of, the singular role overseeing, go-to-market?

Amanda Kahlow:
I think it’s in the next 12 to 24 months.

Mary Shea:
24.

Amanda Kahlow:
I think it’s, like, 12 to 24 months where it starts to become the commonplace. I think some companies are already doing it today, and I think the early adopters, and those are the ones who are going to see, like, the zero-day exploit. Those who do this now, and actually start to rethink their organization, and think of AI as a top-down mandate, not a bottoms-up. Everyone is talking about bottoms-up. I need to train my ICs on how to use AI to make my tasks efficient. everything I do more efficient, but that actually isn’t going to move the needle. You don’t want them doing the same jobs. We need to tell them, like, here’s your new job, what you need to do, and here are the AI tools that are going to help support our buyers across the whole life cycle.

Mary Shea:
Awesome. Yeah, thank you, thank you for that. Yeah. Sandra, let’s go back to you. So, your team’s deploying, you know. a collection of agents across, the life cycle, right? How do you… how do you think about, stitching together all these multiple agents and… and creating continuity across the buyer’s journey, both from a technical as well as a business standpoint?

Zander Pease:
Yeah, it’s a great question, and so just as a refresher, we do agents across chat, email, voice, text, and then we have product demos as well. And so I think initially this touches on one of the earlier questions, is it, like, a single agent, or is it, like, a swarm of agents? And I think I kind of agree with Amanda, like, under the hood, it’s a lot of different agents doing this work, and how you present it from a marketing.

Mary Shea:
Right.

Zander Pease:
can differ between being one big agent or a lot of super agents, sorry, a lot of smaller agents. So you need to build a system that has just a single continuous thread across all those channels, so that your buyer journey is consistent, and every agent that’s jumping into a touchpoint has a complete knowledge of everything that’s already happened with that buyer. And then, you know, as an aside, have that all work with your CRM. And then your question around how do you think about handoff? Again, we have this thesis that every customer is going to have a really different go-to-market motion, and that handoff should be defined by the customer. And so we let customers define signals where you want to hand off from an agent to a rep, or vice versa. A common example might be, okay, a lot of customers think about speed to lead, they want to have reps respond to high-value leads during business hours, but they’ve realized that having an agent respond after business hours, that really immediacy of touchpoint is actually better for their business model than waiting for the person, the next morning to respond. And that’s one of, like, potentially infinite examples, but you have to define that, within an agent somewhere and delegate that responsibility to make the decision whether to act immediately or whether to hand off to that sales rep, you know, whenever it may be.

Mary Shea:
Got it, got it. Let’s move to Depender, and again, you’re talking about coordinated systems across the full funnel. Where do you see organizations, or your customers in particular, get this wrong when they start deploying multiple agents and multiple, activities start getting automated?

Deepinder Singh Dhingra:
Yeah, first of all, great, great conversation. I’m actually learning quite a lot. I think when we talk about… when Amanda talked about, you know, being, like, there’s gonna be this whole end-to-end GTM motion, we completely agree, that’s what, kind of, most of our customers that we work with are kind of trying to achieve a full funnel approach. to how they think about their GTM motion, rather than, this is marketing, and this is SGR, BDR, and this is AE, and customer success, right? I think Xander mentioned that as well. So, one of the biggest problems to achieving that is, in the past, there were, like, these 23 or 24 or 25 SaaS tools that people invested in, right from the CRM to the marketing automation systems, to the sales automation outreach tools, to ABM intelligence tools. to chat engagement tools, etc, right? And what that led to is, right, while each of these tools is amazing in and of themselves, what that led to is fragmented GTM. Right? So, although all of us and our VC ecosystem kept investing in those tools, GTM efficiency kept coming down. Right? Ironically, to Amanda’s point, like, you know, it went from $1.5 for a dollar per AR to about $3.5, approximately, right? And it was exactly because all of us invested in point solutions. Now what’s happening is those point solutions are getting… getting replaced by point agents, right? And and now… but it’s not 23 tools, it’s… or point agents, it’s actually 300. point agents, right? So now you have this problem, you know, exponentially increased, right? Because now you have, like, 300 agents that you’re trying to manage, that you’re trying to coordinate, they’re trying to make sure that they all are feeding off the same context, they have the right guardrails, they take the right persona. Right? Because you want to understand… each agent needs to understand not only all the touchpoints. Each agent needs to understand the goal of the GTA motion, the persona, the brand tone, right? What are the handoffs, right? At what point they hand off to each other, and that is the nature of the problem. So, our perspective is that instead of investing in one agent at a time, invest in a system so that you can deploy a team of agents. And our perspective, our vantage point is that we are working with companies that have complex GTA motions, right? Multi-product, multi-channel, multi-segment, multi-hybrid, it could be ABM, inbound, outbound, product-led, marketing-led, sales-led, all operating at the same time. There is no enterprise or even upper-mid market company today that is a simple account-based motion. Right? Or that’s a simple PLG, or that’s a simple outbound, right? They are all hybrid, so managing this complex team of agents. within that context, gets even tougher and tougher. So my only point of disagreement is I don’t think there are superhumans that’ll be able to manage that, right? It’s going to be… there’ll be a superhuman only if there’s a human, right? So there’s going to be agent teammates, and there’s going to be human teammates, right? I don’t believe for complex GTM motors, and I, because I’m saying this because I’ve worked with Fortune 500 and Fortune 100 companies. Before I started RevShore, and I don’t believe that there’s a superhuman agent, either one, or a team of agents that’ll be able to do this. There’ll be superhuman agents only if they’re human agents. and they are AI agents working together, and so that’s the most important thing to understand here, right? The agent of the superhuman agent needs to take the persona of the human agents and understand the human agent’s goal, right? So if it’s an SDR persona, where did that come from? That came from a human? thought process, right? If there’s an AE that… which, you know, which I don’t know why, if a SDR can be replaced, why an AE can be replaced. I’m not saying that an SDR can be replaced, and I’m not saying an AE can be replaced, but the same logical conclusion does not go for AEs, you know, and why it can’t go for SDR, so that I don’t understand, right? Because if you have to kind of, you know, there has to be, like, as if the SDR has no no creativity left, and there’s no human aspect in the SDR, I don’t believe that. So, our perspective is that you need a coordinated system of agents on a shared context layer, sharing the same guardrails, and sharing the same, same set of objectives of the GTO motion, right? And that’s how, you know, you need to start. Rather than start one agent by agent, one superhuman agent from one place, another superhuman agent from another place, will not work. For most companies.

Mary Shea:
Okay, so I, okay, I’m gonna talk, I’m gonna talk today, because I see a point, counterpoint,

Deepinder Singh Dhingra:
Yeah, so I waited, like. I waited 30 minutes, so I had to kind of come with a big bang.

Mary Shea:
You have to come with a big bang, because, everyone has a big bang here on this, but one second, Amanda. So, I’m gonna get, I’m gonna get to you, I promise. So help me understand, just, just a couple, couple, couple of things that are, that sort of… popped out at me when you were speaking. One is, how long, like, what’s the process to educate these three? Say I have 300 agents in my go-to-market organization. What does the process look like to educate them, and what’s the timeline, and how does that get done? Real quickly, if you can, I’m just super.

Deepinder Singh Dhingra:
Yeah, so if you have a good platform that you can customize and configure agents for every stage of the buyer’s journey, for every type of motion, for every type of touchpoint in that journey, right, it could take from days to a couple of weeks, to kind of, you know, configure each agent. Because each agent essentially is sharing out of the same context, on the same platform, with the same set of guardrails. Right? And I do believe that for the enterprise, you have to be able to configure the system of handoffs, so I think someone mentioned that, I’m sorry, you know, if I get the name wrong. You have to be able to configure the guardrails enterprise by enterprise, right? Because enterprise complexity is what most simplistic superhuman agents don’t understand. Right? They’re not built for that, right? You can feed a bunch of 10,000 case studies to superior intelligence, but that is a simple one task, right? Enterprise complexity, where you have trade shows and webinars, and you have prospecting motion, and you have inbound and warm outbound motion, etc. By the way, in the enterprise, in the enterprise. cold outbound, warm inbound, as we are prospecting is only 20% of the GTM ocean, believe it or not, right? Right? So you need a system of agents, and, you know, if you have the right platform, you can be able to do it in a matter of days to a matter of weeks. Couple of weeks.

Mary Shea:
the speed is amazing. I think what I’m also hearing you say is that the human has a role in the go-to-market organization, but I think it’s a very different role. They’re very different roles than what they had in the past, which I think could actually be a whole other session in itself, but I want to be really fair to Amanda. I know you’ve got a different perspective here, and you’ve been very patient, so please, please let us have it.

Amanda Kahlow:
Well, Deepinder, I would love to put a superhuman to work for you to show you that it is enterprise-ready, and it could go all the way.

Deepinder Singh Dhingra:
I’m an SMB. I’m an SMB.

Amanda Kahlow:
I understand.

Deepinder Singh Dhingra:
It might work for me.

Amanda Kahlow:
65?

Deepinder Singh Dhingra:
Enterprise.

Amanda Kahlow:
5M.

Deepinder Singh Dhingra:
It wouldn’t work for enterprises. It does work.

Amanda Kahlow:
We have 65 enterprise customers, some of the biggest logos, and I promise you they’re not.

Deepinder Singh Dhingra:
Which part of the GTA motion is it implying? follow the GTA Motion.

Amanda Kahlow:
entire.

Deepinder Singh Dhingra:
part of the detail motel.

Amanda Kahlow:
I’m saying the entire motion. I’m saying it does everything from inbound to.

Deepinder Singh Dhingra:
See it when I believe it.

Mary Shea:
I’m not here to prove it to you, but… All right, you both had your moment, so I’m gonna, I’m gonna rein this in, and, that, that, that was really fun, and, and dynamic. But let’s move on a little bit. Ashton, we have your CEO has joined us here, so I don’t know, Nulu, if you’re prepared to speak at all, or did you want to say a few words?

Nilou Salehi:
Hello, unfortunately, I’m not in a position or a place where I can contribute well, I apologize. I’ll let Ashton take this one. But great to meet everyone.

Mary Shea:
Yeah, thanks, thanks for joining, we just so appreciate it. Anyway, let’s, let’s move on and, really start to think about, really measuring progress. So, Amanda, you talked a little bit about, sort of, I think we’re thinking, you know, whatever humans are left in the go-to-market organization, their roles are going to look very different. But I think what you’re saying is the metrics are going to be the same. How are you all thinking about metrics with agents involved in the process, and are there micro-metrics that are important in addition to the late-stage funnel types of metrics?

Amanda Kahlow:
I think there’s two types of metrics. There’s, like, looking at your revenue. At the end of the day, we’re all… we all want to measure to revenue, but I think some of the top of funnel measurements will collapse. Like, looking at just, like, booking meetings and pushing to an SDR and MQLs, those things will… those will go away, and we’ll be looking at sales qualified and accepted leads. And revenue, deal cycles, ACV, like, we have one enterprise company of the largest business social networking company, said we increased their revenue by 2X. And shorten their sales cycle by 23 days. So those types of things. But then, on the flip side, I have other companies, like a very large CRM company, that is measuring this by capacity of humans. They said it would have taken 69… sorry, 89 SDRs and 17 sales engineers to do the job of their one superhuman. Across the life cycle of their business. So, how many human replacements can you replace? If you look at the conversations that she’s having, the complexity, the demos that she’s giving, that would have been a sales engineer job in order to do what she did, and so they were able to quantify the human capacity replace as well as revenue. So, there’s two angles to look at this.

Mary Shea:
Got it. So, we all know that the technology is moving faster than any other new technologies in any of our lifetimes right now. Like, I can’t even… believe what I was doing 6 months ago versus what I’m doing today, with some of the activities that I do day-to-day in my work world. Like all of you, I have a very, very small team, we all have our own agents, we have collective agents, and we’re moving so fast. This begs the question of, what about, you know, the human implications? How… in all of the work that you are doing, you’re all selling into sales leaders or CEOs, how are you finding the human receptivity to some of these changes that obviously are going to bring incredible customer experiences, profit margins to the likes we’ve never seen before, right? We get that. But how is the friction on the front lines with what the human implications are, and the implications that I’m really taking away from this conversation are there is going to be a hell of a lot less go-to-market people But there are going to be go-to-market people, but they’re going to do things fundamentally different than what they’ve done before. I’d love to hear how these conversations are going with your senior-level customers. Anyone want to jump in?

Amanda Kahlow:
I’m happy to, but I feel like I’ve been talking a lot, so I was trying to.

Deepinder Singh Dhingra:
Yeah, I can just take a… very quickly, our perspective and what we are seeing is that the role is shifting from an execution role to more of defining, designing? Yeah. And monitoring. And becoming the control plane? Yeah.

Mary Shea:
Yeah, edu… so, defining, educate… you know, designing, educating, and then the control operator.

Deepinder Singh Dhingra:
Exactly.

Mary Shea:
And most people didn’t… that I know didn’t go into sales to do that, so, are you… are your companies, your customer companies looking for going to, you know, Chicago booths and the top-tier business schools and just getting, like, super smart operators who used to want to go into finance and now putting them into sales jobs? Like… Amanda, what’s going on in sort of these human conversations that you have, and how are people approaching this massive change?

Amanda Kahlow:
I mean, I think it’s… it’s hard. Like, of course.

Mary Shea:
Yes.

Amanda Kahlow:
So, and it’s, you know, I just got off the phone this morning with the CEO of a public company, and… talking to this person, you know, saying, we want to do this, and we’re going to do this, we’ve already proven it out, that it can happen, but the new… like, navigating the org and the complexity of telling people, like, and, you know, the human element, that this job that you have today isn’t here anymore, and trying to find the right fit for the next job, it’s easier with smaller companies, and it’s harder with.

Mary Shea:
Right.

Amanda Kahlow:
You know, and they have… but there’s massive mandates at the board level that this…

Mary Shea:
Right.

Amanda Kahlow:
So we don’t have a choice. It’s not… we cannot use the same playbooks, because they don’t work, because growth is more expensive, and growth rates are shrinking. So there’s… there’s no other option, and that’s best for, like, at the end of the day, they’re all shareholders of the company, the employees, that’s the way I look at it, is our employees are shareholders, and so… When you look at it from that perspective, we all, like, have the same interests in mind. But I don’t think… I don’t think there’s a magic bullet or an easy answer. I think it’s just there is… there are some emotional things and hard aspects that are happening. And then there’s Tiger teams. Like, a lot of our… these companies are coming up with tiger teams to say, like, how are we going to message this to the company? And almost… We’re gonna do this, but we’re not gonna tell the people right away that it’s coming for their job, so it’s.

Mary Shea:
Right.

Amanda Kahlow:
it, but they’re not being told, and so there’s a disconnect. It’s really tough, and I don’t think there’s an easy answer.

Mary Shea:
Yeah.

Amanda Kahlow:
But I actually think being honest and head-on is the right way to attack it.

Mary Shea:
Yeah, it doesn’t surprise me, and I agree as well. I think, that being honest and transparent, like, people know what’s happening anyway, and if you aren’t honest, they’re going to make up You know, what they think is gonna happen. So, you know, this is a little bit outside this session’s topic area, but just because I have a love for the sales profession, I would love, like, each of you, from your own perspectives, what advice would you give, you know, a mid-career AE at, you know, an SMB or a publicly traded company. Sandra, do you want to start? I know this isn’t…

Zander Pease:
Yeah, start.

Mary Shea:
Don’t beat each other’s…

Zander Pease:
one.

Mary Shea:
I’m sorry, that was kind of mean, but I.

Zander Pease:
No, no, no, no.

Mary Shea:
You’ve got a lot of perspectives.

Zander Pease:
I think about this all day long. So I think if that is your profession, and that’s your love in life, move towards where the human touch is going to matter, continue to matter the most. So again, like, we believe that sales can be totally automated where it should be automated. That tends to be, lower value sales, right? Lower value touchpoints. If your skill in life is working on million-dollar enterprise contracts, dealing with stakeholder alignment, having 30 calls over a year to get something over the line, that’s the last place you’re gonna be, like, negatively impacted. And AI can help you, work on that job in a way that, you know, AI might hurt you in a career that’s in a very different type of sales role.

Mary Shea:
Got it, got it. Depender, you know, you and I, I think, are thinking, like, fundamentally different roles. Like, what are your thoughts?

Deepinder Singh Dhingra:
Yeah, so I think that two things for A’s. One is be very comfortable with using agents and agent-take AI for different tasks that you were doing manually. Whether that’s account research, or lead research, etc, trying to understand the… the needs and pain points a bit more, right? And then some of the follow-up, etc. will get automated, and you’ll be using more and more of agents that help you. On the other end is become an expert in your customer. Become an export in your industry. Right? Become a registered in, like Alexander mentioned, the stakeholder alignment and the software aspects, right? Because that’s, you know, that’s what’ll differentiate you.

Mary Shea:
Thank you, thank you for that. Well, let’s… let’s go back to, sort of, the compliance and the monitoring and control aspect of, what we’re talking about here. So, Noam, I’m gonna, turn… turn the camera over to you, but… or the microphone over to you, but… You know, for… for large organizations, any organization, really, what should… what should, the technical team be monitoring in real time to just make sure, these agents don’t go off the rails? Like, you know, and what signals would you be looking for to see a system that’s drifting, or becoming unsafe, or some sort of failure? How do you think about that topic?

Noam Schwartz:
It’s actually super straightforward and simple, and and I… by the way, the discussion that Amanda and Divine there had about, like, is this real? Is it in production or not? Now, we protect some of the… we protect Fortune 500, we protect, the models of, MAG7, and we also protect, regular enterprises with, with, like. Tens of millions of dollars of, of revenue, and… It’s so… it’s really based on how… what kind of sale is it, what kind of product, how complicated, what’s the average sales price, Who’s the… who’s the audience? So we see everything. Like, we see a lot of skepticism, but we see a lot of implementation, that is completely autonomous, but I can’t… but still, the technology Isn’t there for most people, for mostly… for the complicated process. And… but it can be there, because it’s all about trust. Why it’s not there? Because they can’t trust the technology, because they tested it. And it didn’t draw them the results that they wanted. And the way that A smart, accountable organization is implementing guardrails, is implementing the right protections, is effectively connects all of their agents to an independent guardrail. that can actually understand the policy of the organization, and independent doesn’t mean to have, like, a separate vendor, but an independent concept, like a feature in a platform. or something that you can set up independently, not something that the vendor that is selling you the product is building for you. You need to build it yourself. You need to be accountable to the policy, and then you need to test it. So once you’re implementing a guardrail, and you’re seeing that it’s covering everything the model is doing.

Zander Pease:
Everything.

Noam Schwartz:
is coming into the agent, everything that is coming outside of the agent, every tool calling, because the agents can call tools or other agents, every access to the database, like a complete auditable log that you can look into each and every mistake, and you can patch,

Zander Pease:
Issues.

Noam Schwartz:
you can block specific vulnerable things, and you can say, like, okay, this is, like, too complicated, I don’t want the agent to go here. We’ll re… we’ll revisit it next time. So have this guardrail, layer that you understand. This is super important, not like a black box. Something that you understand how it works, and simultaneously, there’s a concept called red tiering. It’s like pen testing to the organization, but it’s for the agent. You take your agent. you put it, like, in the gym, like, and you start, like, you start, like, running it around. And you go to every single option that the… every single scenario that the agent can, run into. And I’m talking about hundreds of thousands of scenarios. And then you know when something… that something can happen. Right now, most people are not really managing risk, they’re kind of, like, accepting the risk. Like, we know that something can happen, maybe in the long tail, I don’t know, who’s going to go after me. there’s this kind of, like, very naive approach to everything that is going on, especially those who just, like, want to progress, which I’m encouraging. But in… when… when agents meet the real life, then people start facing the real consequences of not testing. And that’s, again, that’s the way to do that. Guardrails. Red teaming, testing, and then you know that when you’re in production, you’re actually managing a risk. You’re not just accepting it.

Mary Shea:
Got it. Thank you for that. Could I add to that?

Zander Pease:
Quickly?

Mary Shea:
Very quickly, because I want to… we’ve just got a few minutes left, and I want to do a quick lightning round.

Zander Pease:
Yeah, so, because as a technical leader of, like, working with a lot of non-technical folks, I agree with what Noam says, but I think it might help the audience to, like, really define what some of these terms of art mean, because we’re using, like, guardrail a lot, for example. So just, like, very simply, we have a lot of customers that go live with an agent, and then they expect someone to have to read every single conversation that an agent does. And one of the concepts here is usually called LLMS Judge. But just, like, very simply, you can deploy an agent out in the wild, and then have another AI look at every single conversation for you, and pull out, like, the 1 or 2% that your people need to look at. And that’s very cost-effective, it’s not prohibitive to your business, it can really cut down on the complexity that you might have in your head about what it’s like to deploy agents. And that leads into this second concept. human in the loop, right? So when you actually do have something that you need to review. How do you have a platform that makes it really easy for someone on your team to review? And because these are go-to-market people reviewing it, how do you have them review it in a way that then shifts it over to maybe a RevOps person, or maybe you have an engineer on your team that can take it and help the model improve over time? And so we’ve touched on that in a few different ways, but I think it’s helpful to just sort of explain exactly how that system works in layman’s terms, because a lot of this is actually a lot less scary and a lot less work than you might think it is, even once you have a huge number of agentic conversations out in the wild.

Mary Shea:
So thank you for that framing. All right, we’ve got 2 minutes left, and hopefully Julia will give us a little bit of a break if we go slightly over, but I’d love to hear, and let’s start with you, Ashton, from everyone. What’s gonna separate the companies that successfully manage AI as a true teammate, as a workforce, from those who don’t? And Why don’t you go ahead, Ashton?

afshinnikzad:
Yeah, I agree with much of what Noam said, the importance of taking, evaluations seriously, having a systematic way of evaluations. But I also want to add to that, you know, not everything there is super straightforward. For example, when it comes to taking actions, which is essentially the, you know, where the value is in agent deployments, we want them because of execution and actions, right? Think about a POC example I mentioned, an agent who is generating a POC draft. As context grows inside an opportunity, it becomes more and more difficult to track. why that POC doc was generated. It’s… the models are like a black box, so we need to build systems, right, to track that and make that explainable, and that part is not straightforward. So I think two things I would say. Taking evaluations seriously, and having systematic approaches For how their agents that they deploy, make decisions. In a cross, Don Lock for us was the research that we had done at UC Berkeley and USC, and basically turned that context to execution layer based off on that. But yeah, those are the two principles I would advise.

Mary Shea:
Thank you, thank you for that. Amanda, what’s going to separate the winners from the losers here, in your view?

Amanda Kahlow:
think about AI from a top-down approach versus a bottom-up, so that’s the number one thing. So instead of telling all your ICs to be better with AI, you need to think about where is there… to start, where is there no business model today to put a human and put something like a superhuman into your process and into your models? And also to think… I think we need to start moving away from this world of experimentation. We don’t do any pilots. experimentation error is over. The stuff works, that you need to really be giving it, like, real enterprise attention, to make it successful.

Mary Shea:
Perfect. Depender.

Deepinder Singh Dhingra:
Yeah, I’d like to add great points earlier. One is have a shared context. Make sure all of your agents Are operating out of the shared context, rather than going rogue. and doing their own thing. I think that’s a very important… Aspect, shared context, and the coordination between the agents.

Mary Shea:
Beautiful. Xander and Noam, can you each weigh in super quickly? I see Julia, so I know she’s ready to tee up the next group.

Zander Pease:
Go ahead, no.

Noam Schwartz:
So fixing security and safety before something happens would actually help the adoption and bringing the ROI, like, as soon as possible. That’s the most important piece. Like, shift left, this whole, responsibility, make sure that you’re prepared, because AI is a long-tail problem. It’s not a deterministic issue, and the whole way that an agent behaves, something will happen. And if you’re prepared, you will win for the long term.

Mary Shea:
Got it.

Zander Pease:
Endurance.

Mary Shea:
Sander.

Zander Pease:
Very quickly, we think AI makes sales into a science, so all of a sudden, as a sales leader, you can start to A-B test every single part of the interactions that your customers have with your agentic personas. And in a platform that helps you create those AP tests, that thinks of those AP tests, that runs them automatically over time, they’re gonna eke out more and more gains in your go-to-market funnel. Than you could have done with the human sales team.

Mary Shea:
Alright, I love it, and that’s a wrap. This has been amazing, I have learned so much, and I, expect to hear from everyone on this panel, because, I want to learn more about your companies and potentially even use, your solution at Meerkat. So, Julia, back to you.

Julia Nimchinski:
Thank you so much, our panelists, and thank you so much, Mary, that hour flew by. Mary, what’s the best way to support you? What’s the latest and greatest with Meerkat?

Mary Shea:
Yeah, I mean, it’s super exciting right now. At Meerkat, we’ve got, multiple products that are now operational, in beta, and we’ve got the inklings of a commercial pipeline, so we’re moving forward quickly, and we’re doing both the horizontal and vertical approach. PLG, and also some selling. So, you can find me on Meerkat, TryMeerkat.ai, and you can find me on LinkedIn. I answer every direct message, believe it or not, even when I’m getting pitched.

Julia Nimchinski:
Thank you so much again.