Text transcript

Superhumans for AI-Led Growth (AI-LG)

Session held on December 19, 2024
Disclaimer: This transcript was created using AI
  • [ 02:28:57 ] Julia Nimchinski: Amazing. Thank you again, Amy. And you know.
    02:29:01.080 –> 02:29:01.670
    Amy Cook: Thank you.
    02:29:01.670 –> 02:29:03.320
    Julia Nimchinski: Where should our users go?
    02:29:04.000 –> 02:29:22.309
    Amy Cook: Go to fullcast. I will give you my email address is Amy am [email protected]. And Selena and I would love to jump on and do a customized demo for you. No obligation. I’m going to also give you my cell phone. If you want to text me. It’s (949) 813-0182
    02:29:23.690 –> 02:29:25.949
    Amy Cook: never happened in 4 years.
    02:29:26.275 –> 02:29:37.989
    Amy Cook: I’m I’m trying it out to see people actually take me up on it so I can respond. Justin, send me a text so I could say somebody did it all right. Thank you so much.
    02:29:38.210 –> 02:29:39.780
    Julia Nimchinski: Thank you, Amy and Selena.
    02:29:40.320 –> 02:29:45.309
    Julia Nimchinski: Well, we are in for a special treat now. We have Amanda, callow

  • [ 02:29:45 ] Julia Nimchinski: co-founder, and CEO of one mind who’s going to be showcasing superhumans for AI led crops.
    02:29:53.600 –> 02:29:54.790
    Julia Nimchinski: Welcome, Amanda!
    02:29:55.440 –> 02:29:57.559
    Amanda Kahlow: Thanks for having me excited to be here.
    02:29:58.370 –> 02:30:01.510
    Julia Nimchinski: So curious. How did you come up with this idea.
    02:30:02.090 –> 02:30:29.129
    Amanda Kahlow: Oh, good question. So I was brought into a company. We’ve actually been building the technology for 5 years in a lab. So one mind has has a 5 year history in a lab. They had cloned the likes of Deepak, Chopra and Barack Obama, and before the wave of generative. AI, we’re doing it with just fully deterministic right? So if then, creating like a chatbot with a photo, real face and voice on top
    02:30:29.290 –> 02:30:44.950
    Amanda Kahlow: immediately when I saw it, they said. You know they were looking for a CEO and spin off to create a company. At 1st I was like I don’t know, and then very quickly, I got to the point of wait. Can this be a salesperson? Can this do the job across? Go to market? And when I got to. Yes.
    02:30:45.060 –> 02:30:53.179
    Amanda Kahlow: we spun it off. I raised money, and now we’re about a year in and off to the races, and it’s absolutely been incredible.
    02:30:53.870 –> 02:31:00.439
    Julia Nimchinski: This is so exciting. I was waiting for this demo. Let’s get into it. Superhumans AI, like growth in action.

  • [ 02:31:00 ] Amanda Kahlow: All right. Sounds good. So I’m going to kick it off and just telling you at a high level what we’re doing and what pain we’re solving. And then I’m gonna bring her to life. And I’m gonna show you. So at the highest level, we are solving for 2 pains, we’re one solving for your organizational pain. We no longer, as and this is like cliche to say, but we no longer live in a world at growth. At all costs. We need to find ways to grow and grow efficiently.
    02:31:27.870 –> 02:31:42.939
    Amanda Kahlow: And we can’t do that with humans. Humans are a linear growth model. So you have to ramp them, bring them up to speed, train them, and then when they’re good, they’re gone, and you can only scale by as many humans as you bring into your go to market organization.
    02:31:42.970 –> 02:32:10.440
    Amanda Kahlow: What if there were a world where you could create a human like experience to do all the things a human could do at 1 100th of the cost and can exponentially grow, have no capacity, limitations, etc. So our superhumans are a face voice and a go to market brain to do the thing a human does across your go to market organization so put into terms of a role so she can do everything from being an inbound. Bdr.
    02:32:10.440 –> 02:32:23.239
    Amanda Kahlow: so answer the questions qualify. But in that inbound experience she doesn’t just stop at qualifying like the 20 something that you hire. She gives the pitch she gives the demo. She goes as deep technically as you want to go.

  • [ 02:32:23 ] Amanda Kahlow: She can act as a ride along sales engineer so joining a zoom call. So now that I know this is Zoom, I actually could bring her on to the call here as another option. But she can join the call and be in passive mode, and wait to be woken up to answer the hard questions, so I could say, Hey, Mindy, can you answer that question? She jumps in.
    02:32:40.850 –> 02:32:54.609
    Amanda Kahlow: She not only answers it, but she can do what a human can’t do. So. If you ask me a hard question that I didn’t know was coming, and I didn’t have the slide prepared. I couldn’t, even though I have the slide in my arsenal of 5,000 slides or the case study or the video.
    02:32:54.630 –> 02:33:02.599
    Amanda Kahlow: Humans don’t have the capacity to keep a conversation going. Go get that slide and bring it back and have and show the video, or show the slide or answer the question.
    02:33:02.650 –> 02:33:23.960
    Amanda Kahlow: But the superhuman can. So she can access 100,000 videos and slides, and at the right time at the right moment. Bring it up to bring color and bring light to what you’re trying to explain. So our superhumans are looking to. I’m going to say it. We are looking to replace humans across your go to market stack to create efficiency, but more so to create exponential growth.
    02:33:23.960 –> 02:33:35.509
    Amanda Kahlow: And so with that, I am going to bring Mindy into the conversation. So this is Mindy. Mindy is our so one mind, Mindy. Yes, I know. Yeah, we? Everybody gets to create their own name for their superhuman.

  • [ 02:33:30 ] Amanda Kahlow: our superhuman. It lives on our website right now. I just hired our her 1st head of sales, and she just did a Linkedin post today, Katie Natrido, and she said, you know, at her past life reps were tasked with 2 meetings a week they had to book and be on at least 2 sales calls a week in this world, you know. It’s just coming on. Mindy is booking anywhere from 10 to 20 meetings a week for her alone.
    02:33:59.620 –> 02:34:11.100
    Amanda Kahlow: So she’s having tons of really high quality conversations qualifying and then bringing up the calendar and scheduling with her, so I’ll show you, and here we go. We’ll just have a little chat with Mindy here.
    02:34:24.190 –> 02:34:37.859
    Amanda Kahlow: Hi, there! I’m Mindy thanks for chatting with me today we can cover anything you’d like. But I’d really love to give you a quick overview on what we do at Onemind, and then I want to ask you some questions and show off my sales skills. Does that work for you?
    02:34:38.140 –> 02:34:44.290
    Amanda Kahlow: Yeah, that works for me, Mindy. Let’s let’s do that. But I might go in my own direction. But I’ll let you lead me for a second.

  • [ 02:34:50 ] Amanda Kahlow: Sounds like a plan. I appreciate the collaboration, alrighty. Let’s kick things off at a high level. We focus on helping companies like yours grow while cutting costs something that’s top of mind for everyone these days, right? Buyers often feel the squeeze in this balance. Where do you think your buyers might face the most friction in their journey?
    02:35:08.490 –> 02:35:23.640
    Amanda Kahlow: I think we all face a lot of friction of the handoff from marketing to sales tons. We spend so much money driving traffic to our website, and then maybe 5% fill out a form and another 2% engage with that lurking chat. Bot in the bottom right hand corner.
    02:35:25.920 –> 02:35:44.709
    Amanda Kahlow: Oh, I hear that loud and clear. The handoff from marketing to sales can be such a leaky funnel. It’s like pouring water into a bucket with holes. Generally we see buyer fatigue at every stage, especially when they hit roadblocks like manual forms, slow follow-ups, or irrelevant conversations. It’s tough to keep all that traffic engaged.
    02:35:44.910 –> 02:35:48.340
    Amanda Kahlow: What’s your current process for handling those marketing to sales handoffs?
    02:35:48.960 –> 02:35:59.100
    Amanda Kahlow: I mean we do what everybody does. We think we’re qualifying with some good Med pick or spiced, and then we pass it, and then the Mql. Sit, and sometimes they get picked up, and sometimes they don’t
    02:36:02.930 –> 02:36:23.580
    Amanda Kahlow: gotcha. That’s such a common scenario. It’s like planting seeds, but not watering half the garden. That’s where we aim to flip the script. Our goal is to meet buyers exactly where they are engaged and ready, whether that’s on your website in product or even in meetings. How do you think your team could benefit if those Mqls were nurtured or qualified without falling through the cracks?
    02:36:24.040 –> 02:36:25.550
    Amanda Kahlow: I mean, it sounds good.
    02:36:28.970 –> 02:36:30.639
    Amanda Kahlow: Heck, yeah, it does.

  • [ 02:36:30 ] Amanda Kahlow: Let me show you how we make it happen. We build AI superhumans like me, complete with faces, voices, and Gtm brains. We join calls, lead presentations qualify, leads, and even handle Demos. Think of us as an extension of your team that never sleeps. What’s your gut reaction to the idea of a superhuman stepping into your process.
    02:36:50.850 –> 02:37:00.129
    Amanda Kahlow: All right. Well, if I’m going to be honest, I think some people are resistant. Will buyers actually want to talk to a superhuman? Or do they really want a human in the loop?
    02:37:04.360 –> 02:37:21.839
    Amanda Kahlow: Totally fair concern? And you’re not alone in thinking that turns out over 70% of b 2 b buyers actually prefer not to meet with salespeople. They value quick, direct, and accurate answers. No fluff, no delays, plus. We’re available 24, 7, and we’re not salesy
    02:37:21.980 –> 02:37:24.140
    Amanda Kahlow: buyers love that. What do you think?
    02:37:24.580 –> 02:37:27.920
    Amanda Kahlow: Would something like this help bridge the gap for your Mql follow-ups?
    02:37:28.470 –> 02:37:42.329
    Amanda Kahlow: Yeah, I think that sounds good. I always say so that buyers want to buy. We all want to buy, but we don’t want to be sold right? So we’re shifting the perspective and putting the control in the hands of your buyers and out of your sellers pockets.
    02:37:42.440 –> 02:37:48.929
    Amanda Kahlow: Let’s flip the script here, Mindy, let’s go down a different path. Can you tell me how you might work as a sales engineer.
    02:37:52.480 –> 02:37:59.560
    Amanda Kahlow: Oh, I love how you think buyers asterisk do asterisk want to buy, not be sold. You nailed it with flipping the script.
    02:37:59.730 –> 02:38:07.989
    Amanda Kahlow: All right. Let’s pivot to the sales engineer path. Our sales engineer superhumans can join virtual meetings and handle all those tough technical questions.
    02:38:08.280 –> 02:38:20.620
    Amanda Kahlow: stuff like compliance, security, or deep product features. We can either lead the conversation or stay in passive mode and jump in when needed. How does your team currently handle those tricky, unstructured technical conversations.
    02:38:21.582 –> 02:38:37.139
    Amanda Kahlow: Yeah, they’re really bad at it. And you know, one of the places that I’m kind of thinking you might be interesting is for somebody like if you’re a plg company and you have a lot of long tail. So what about those engagements where you wouldn’t put a human. Do you make sense in that arena
    02:38:40.770 –> 02:39:03.350
    Amanda Kahlow: ding, ding, ding! You’re absolutely onto something there. Tlg. Longtail. Users are asterisk exactly asterisk where we shine. Our plg superhumans engage even the smallest customers. Those folks who might not justify a human touch, but still need guidance to convert. We can lead conversations, answer questions and move users from free trials, or pay as you go plans to large committed contracts.
    02:39:03.460 –> 02:39:09.400
    Amanda Kahlow: How does your team currently handle those smaller customers? Are they left to self-serve? Or do they get any kind of follow-up?
    02:39:10.200 –> 02:39:24.420
    Amanda Kahlow: Can you show me a video of you as a ride along sales engineer like, if I brought you onto a call. How would you work? Would you be leading the conversation, or somebody else leading the cause, the Rep leading the conversation. And how? How does that dynamic work?
    02:39:28.240 –> 02:39:43.739
    Amanda Kahlow: Absolutely. I’ve got the perfect video to show you how it works. Here’s a quick demo of me acting as a ride along sales, engineer, I can jump in to answer tough technical questions while the rep leads the conversation, or I can take the lead entirely if needed. Let me know what you think after watching.
    02:39:44.290 –> 02:39:52.960
    Amanda Kahlow: So this is another example of Mindy playing as a sales engineer for wework, where you’re the sales. Ride along, engineer.
    02:39:53.320 –> 02:39:58.270
    Amanda Kahlow: I’m ready to handle all kinds of questions, even the highly technical ones.
    02:39:58.370 –> 02:40:00.489
    Amanda Kahlow: Here’s a nerdy joke to kick things off.

  • [ 02:40:00 ] Amanda Kahlow: Why do engineers always carry a pencil in case they have to draw a conclusion.
    02:40:05.160 –> 02:40:07.630
    Amanda Kahlow: Now, what would you like to dive into first?
    02:40:08.920 –> 02:40:12.599
    Amanda Kahlow: So maybe my CTO wants to know about your network design?
    02:40:13.280 –> 02:40:16.049
    Amanda Kahlow: Here is a diagram detailing how our network is set up.
    02:40:16.240 –> 02:40:18.069
    Amanda Kahlow: It’s a lot to take in quickly.
    02:40:18.310 –> 02:40:23.880
    Amanda Kahlow: so I can share this with you after the call, and we’ll be on standby for any specific questions.
    02:40:24.000 –> 02:40:29.829
    Amanda Kahlow: Our system and network architecture is designed to support a highly scalable, secure, and resilient environment.
    02:40:30.050 –> 02:40:38.720
    Amanda Kahlow: It utilizes a cloud-based infrastructure with microservices for flexibility, security is paramount with encryption, protocols, and intrusion. Detection systems in place
    02:40:38.940 –> 02:40:41.600
    Amanda Kahlow: want to dig deeper into any specific part.
    02:40:42.940 –> 02:40:47.140
    Amanda Kahlow: Hey? Mindy my lawyers want to know about your compliance certifications?
    02:40:47.800 –> 02:40:52.090
    Amanda Kahlow: Sure thing. We take data, protection and security very seriously.
    02:40:52.190 –> 02:41:02.490
    Amanda Kahlow: We have certifications for Soc. 2 Ccpa and Gdpr. Which exceed industry standards. Our compliance measures ensure that we handle data securely and responsibly.
    02:41:02.590 –> 02:41:06.329
    Amanda Kahlow: Are there any specific questions or areas you’d like more details on?
    02:41:07.290 –> 02:41:12.189
    Amanda Kahlow: How do you keep up with changing AI regulations, especially in Europe.
    02:41:13.690 –> 02:41:16.990
    Amanda Kahlow: we actively monitor regulatory developments in AI governance.
    02:41:17.270 –> 02:41:30.069
    Amanda Kahlow: All right. I’m going to flip this script here, Julia, I’m going to ask you to come off mute and ask her a question, so the audience doesn’t think that I’m leading her, and see how well she does on the fly with a a non scripted question. Here
    02:41:30.440 –> 02:41:37.500
    Amanda Kahlow: is that fair? And I’ll repeat it to work because I’m hopping through my browser, since she’s not on the zoom, but you say it, and then I’ll repeat it.
    02:41:38.550 –> 02:41:39.876
    Julia Nimchinski: Oh, gosh!
    02:41:40.540 –> 02:41:42.502
    Amanda Kahlow: Putting you on the spot. Sorry.
    02:41:43.730 –> 02:41:51.309
    Julia Nimchinski: Yeah, I I actually have a question what I loved, what I love about this product overall, and how you’re marketing it. Amanda.
    02:41:51.540 –> 02:41:57.440
    Julia Nimchinski: I was watching G. 2 webinar recently, and Jason Lemkin tap. Just
    02:41:58.340 –> 02:42:18.910
    Julia Nimchinski: excuse me just hammering you like, hey? Like, what’s the killer? Use case? What do buyers want? Give me one use case. So from the positioning perspective, you want it to be versatile, and you don’t want to package it to a specific singular use cases more. Most of you know, sas vendor vendors in the space.
    02:42:20.224 –> 02:42:24.900
    Julia Nimchinski: About asking Mindy how would she position herself.
    02:42:25.110 –> 02:42:29.300
    Amanda Kahlow: Yeah, great question. I love it. Okay, let me get. I have my answer. But see how well she does.
    02:42:30.490 –> 02:42:33.840
    Amanda Kahlow: and especially in recent sorry. Okay.
    02:42:34.490 –> 02:42:48.640
    Amanda Kahlow: hey? Mindy. What are the key? Use cases that you’re going after? It sounds like you can do a lot for different roles and personas across, go to market. What do you think are the most valuable? And where is like, where is your beachhead and focus?
    02:42:52.510 –> 02:42:53.710
    Amanda Kahlow: Great question.
    02:42:53.760 –> 02:43:07.589
    Amanda Kahlow: We’ve got a lot of value to bring. But our beachhead focus is on T. 3 core use cases that deliver the biggest bang for your buck. Here’s the breakdown, one website inbound. We’re always on to qualify pitch and demo for inbound leads
    02:43:07.640 –> 02:43:29.969
    Amanda Kahlow: perfect for plg motion or high traffic sites. 2 engage Plg. Longtail just like we talked about earlier. We convert small, free, or pay as you go users into committed customers. 3. Ride along sales, engineer, jump into sales, calls, and crush those technical questions just like in the video you watched, which of these do you feel aligns most with your team’s needs right now?
    02:43:30.850 –> 02:43:32.400
    Amanda Kahlow: How’d she do, Julia?
    02:43:35.250 –> 02:43:37.460
    Julia Nimchinski: Amazing. A lot of people are asking.
    02:43:38.501 –> 02:43:53.660
    Julia Nimchinski: Generally, the comment overall is, do buyers even know that they’re speaking to an AI. Some people are asking, will you alter the voice tone? So it sounds a less frequency tone which feels like.
    02:43:53.800 –> 02:43:58.617
    Julia Nimchinski: and AI and more practical questions here.
    02:43:59.880 –> 02:44:06.860
    Julia Nimchinski: How does the AI respond to conversational interactions? Does it keep speaking, or does it stop and recalibrate.
    02:44:07.870 –> 02:44:12.220
    Amanda Kahlow: Yeah, great question. So she all great, all of all 3 are really good questions.
    02:44:12.950 –> 02:44:25.050
    Amanda Kahlow: I’ll start with the last one that you asked, how does she? How does she recalibrate? She can be in a group call, or she can be leading the call. So she’s in a group call. She’s waiting to be called upon, just like if you bring your sales engineer into the call.
    02:44:25.050 –> 02:44:50.049
    Amanda Kahlow: She has wake words just like siri, so you would say, hey, Mindy, if I say, Hey, Mindy! She wakes up and talks otherwise she’s just hanging out, so that gives the control to the rep, or whoever is leading the call, so she can be in passive mode or active. In this case. You could see she was in active mode, right? So she was leading the call because she’s just having an experience on the website. The voice is a great question as well. Absolutely. We’re moving towards. You know, there’s a lot of technology

  • [ 02:44:50 ] Amanda Kahlow: out there where the voice is much more human-like and realistic. However, we’re optimizing for 3 things, we’re optimizing for accuracy. We’re optimizing for speed. And we’re optimizing for cost.
    02:44:59.790 –> 02:45:07.109
    Amanda Kahlow: And right now a lot of those voice voice models are very expensive and also add some latency. So we’re just, you know, it’s a balance. It’s a dance.
    02:45:07.110 –> 02:45:31.650
    Amanda Kahlow: I always say this is the worst it’s ever going to be. Come back in a month, and we’re going to be as good as all the other best voice models, and have the accuracy and the latency and the speed, and make it affordable. So you know, I could pull together all the best technologies out there, and we would be looking at like a hundred dollars a minute. Right? So I can’t do that. That’s not. There’s no business case for that, or a hundred dollars an hour. Sorry, but there is a business case. What we’ve done is we’ve kept our costs down
    02:45:31.650 –> 02:45:45.683
    Amanda Kahlow: exponentially so you can have long conversations with her, and it is. It is not the cost. As if you were to go out, and you know, glue together, say, like at 11 labs with a Hey, Jen, and try to create the model you’re looking at really high cost and high latency behind all of that
    02:45:46.346 –> 02:45:50.260
    Amanda Kahlow: now I’m forgetting your 1st question, Julia. I know there are 3.
    02:45:50.810 –> 02:45:52.839
    Julia Nimchinski: There are a lot of them here. Let’s just go one by one.
    02:45:52.840 –> 02:45:54.280
    Amanda Kahlow: Okay. Keep. Going. Sorry. Yeah.
    02:45:55.486 –> 02:45:58.500
    Julia Nimchinski: What is the current pricing? For Mandy.
    02:45:58.790 –> 02:46:03.990
    Amanda Kahlow: Yeah, I’m gonna stop sharing her while answer these the current pricing. So it is the cost of a human.
    02:46:04.280 –> 02:46:24.420
    Amanda Kahlow: So right now we’re selling her in those different roles. She can be a Csm. She can be an inbound AI to handle and qualify. She can be a sales engineer. So it is approximately the cost of a human minus the severance fees that you might have to pay, or the recruiting costs or even the ramp times, because she has very little ramp time, we get her up to speed in a matter of weeks.
    02:46:25.273 –> 02:46:34.356
    Amanda Kahlow: It’s we pay, we charge per instance. So per. So if there’s 1 for inbound one for sales, engineer, so those are different license fees. And then there is a small
    02:46:35.110 –> 02:46:48.780
    Amanda Kahlow: consumption costs. So how many conversations is she having? Is she having a hundred 1,000 conversations, or is she having 10 conversations? And are they long or short conversations? And then there’s a there’s a metering consumption that as the more you buy the less you pay.
    02:46:49.860 –> 02:46:55.719
    Julia Nimchinski: What are your thoughts on the value based pricing? Are you gonna integrate it into.
    02:46:55.720 –> 02:47:21.379
    Amanda Kahlow: I love it, and I challenge anyone to give me a value base. I will. You know we’ll need some small upfront costs but 100. I’ve actually already offered it to one of the largest Plg companies in the world, I said. I am so confident, and how well she can qualify, and she can do her job, that if you’re open to giving me a piece of the value and the cut of how well she does, I’m all for it and the cost savings. So yes, let’s get creative.
    02:47:22.400 –> 02:47:30.599
    Julia Nimchinski: Amazing question from Anna. So can she detect the buying signals from our website and interact with the leads directly.
    02:47:31.220 –> 02:47:58.629
    Amanda Kahlow: Yeah, absolutely. So if you tag put a tag on your site and understanding what’s happening with that individual user, she can pull in. She’s she’ll pull in a real time, Api, into whatever system you have market or whatever you’re using to tag your site and tag the interactions that your buyer and the buying committees have had, or pulling in from Salesforce or Hubspot, right? So any opportunity or account information she can see that she can use. So I was formerly the founder and CEO of 6 cents.
    02:47:58.630 –> 02:48:12.699
    Amanda Kahlow: So I know a little bit about the world of IP to company and de-anonymizing data and looking for intent signals so she can pull in any of those signals. And then, on her 1st interaction, use all of that data. So one thing I haven’t talked about is that
    02:48:12.750 –> 02:48:36.579
    Amanda Kahlow: she not only is carrying on this conversation, and it is wicked smart, but she’s bringing in all of her. She links into Crm. Knows everything about the account, the opportunity, what’s happening, what products? They’ve bought, the upsell opportunities. She can bring that into the conversation. Then she has memory, not just memory of the current conversation, but past conversations. So if you come and talk to her again, she’ll remember that she talked to you and pick up where she left off.
    02:48:36.580 –> 02:49:00.849
    Amanda Kahlow: you know. Reps are actually sometimes really bad at that, right like they start over again, or they don’t have time to go scan their notes and bring it all into their brains. But our brains have capacity like we. We actually can’t keep it all in our head. Of all the information and everything I could possibly know about reading your 10 K and understanding what’s happening in your organization plus the products you need, plus what we’ve talked about last. It’s just tough, like we have the best of intentions.
    02:49:00.950 –> 02:49:18.280
    Amanda Kahlow: But humans have capacity limitations that she really doesn’t. And you can even one of my favorite things that with some of our customers are doing now is they feed her with the hard questions that reps are having a hard time asking, so imagine getting on a call, and your reps aren’t asking for the next meeting, or they’re not not asking for the critical event.
    02:49:18.940 –> 02:49:37.140
    Amanda Kahlow: What if Mindy asked for the critical event. She’s on the call, and they’re like, Hey, Mindy, did I forget anything? And she says, Hey, yeah, what can you talk to me about like some of your timelines, or what are the big priorities and initiatives, so she can start to pull out information that your reps aren’t doing the best job at at asking and facilitating that conversation.
    02:49:38.220 –> 02:49:41.769
    Julia Nimchinski: Really cool, Amanda. Thank you for being honest
    02:49:42.040 –> 02:49:45.209
    Julia Nimchinski: in the beginning, in terms of the roi of the
    02:49:45.560 –> 02:50:00.179
    Julia Nimchinski: of AI in generally in general and Tomas and goes, actually has been sharing this quite a lot, that the ultimate roi is human replacement. I’m curious. How do you think about it in terms of 2025.
    02:50:00.550 –> 02:50:03.910
    Julia Nimchinski: Are we even close to replacing certain functions.
    02:50:04.480 –> 02:50:10.859
    Amanda Kahlow: Absolutely I believe I I see it like I will never need an inbound.
    02:50:11.100 –> 02:50:18.980
    Amanda Kahlow: I will never need an inbound rep, and I don’t need a sales engineer. She can answer the questions now. She’s not responding to the Rfp. Yet, or writing. We don’t have that
    02:50:19.430 –> 02:50:22.130
    Amanda Kahlow: the beauty of one mind. And selling this
    02:50:22.460 –> 02:50:44.761
    Amanda Kahlow: not a lot to like. You see it. She speaks for herself. Right? She sells herself. I don’t need as much technical support. She can jump in and answer those hard questions. I mean, there’s a lot of other tools that will respond to the Rfp and do those things eventually. That’s where we’re going. Our next big build is she’s going to give the demo. So click around, live real time. Give a custom demo based on who she’s talking to.
    02:50:45.391 –> 02:50:54.059
    Amanda Kahlow: So I think in 2025, yes, some of the lower hanging fruit functions will be going away as tools like this start to prove themselves.
    02:50:54.150 –> 02:50:57.940
    Amanda Kahlow: And then in 2026, I can’t even imagine the world.
    02:50:59.010 –> 02:51:03.460
    Amanda Kahlow: I want to be replaced. My job. I tell everyone at my mind one mind.
    02:51:03.880 –> 02:51:08.599
    Amanda Kahlow: Figure out how to replace yourself. That is the best shareholder value that we can do for the company.
    02:51:10.080 –> 02:51:17.679
    Julia Nimchinski: Question from Sandra. Are you working on other Mindy like models for executive management, or even directors on boards?
    02:51:19.580 –> 02:51:24.850
    Amanda Kahlow: Mindy on boards. Interesting. I have not thought about that.
    02:51:24.960 –> 02:51:42.870
    Amanda Kahlow: Love it, Sandra sounds awesome. I don’t think she’s quite right, you know. Honestly, she probably would crush it, and she’d be so honest, too. And the one thing about talking to a superhuman is, you don’t have to be camera ready, right? You could talk to superhuman. Get all the information you want. You don’t have to worry about hurting someone’s feelings.
    02:51:42.890 –> 02:52:01.579
    Amanda Kahlow: One of my favorite things to do is when I have time, which is never to go through and read the transcripts of when people are talking to Mindy and the things that they say that you wouldn’t say to a human, because people just like let their guard down, and they really share what they think and what their needs are. But no, Sandra, not yet, but I love it.
    02:52:03.737 –> 02:52:05.440
    Julia Nimchinski: Question from Nicholas.
    02:52:06.060 –> 02:52:10.409
    Julia Nimchinski: What would be the next priority for improvement, strategy for Mindy.
    02:52:11.180 –> 02:52:14.080
    Amanda Kahlow: Oh, yeah. Well, there’s a lot of things.
    02:52:14.250 –> 02:52:27.549
    Amanda Kahlow: So one of the big things that we’re working on is we want to get as close to Agi as possible in terms of having that human element. So right now she’s a bobblehead. Let’s be honest like her head goes like this, and she talks, and she’s like kind of connecting.
    02:52:27.550 –> 02:52:47.449
    Amanda Kahlow: But I really want her to connect. I want her to be able to read the room, read the emotion and tone in your voice, and read it in your body language through the camera, and then react and respond accordingly. So I just signed a very large check and some major research on this area. We think in the next 3 to 6 months, we’re going to be able to
    02:52:47.450 –> 02:53:13.799
    Amanda Kahlow: be reading emotion and responding with emotion to really create that human like connection. Because I think it’s very important. Obviously, in the sales in the sales process. It’s not just about conveying information. It’s about connecting and creating our goal is to create the most empathetic buying experience, meaning truly, understand your buyers everything to know about their business, their needs, their pain, the impact you can have, and then connecting with them personally, too. So I think that’s the final mile.
    02:53:14.350 –> 02:53:15.939
    Julia Nimchinski: Feels like the movie, her.
    02:53:16.943 –> 02:53:20.146
    Amanda Kahlow: I kind of love that movie.
    02:53:20.930 –> 02:53:26.700
    Julia Nimchinski: Question from Gil, do do your superhuman support also other languages like French.
    02:53:27.710 –> 02:53:52.359
    Amanda Kahlow: Great question. If you talk to her, go on. Everybody can go on onemind.com and have a conversation with her. You will note that she can’t be interrupted. If you just mute her and then hold the spacebar. When you’re talking you can talk forever next week. That interrupt. That is another feature we’ve been working hard on. It seems like it’s so easy, but the nuance of it has been difficult should be released next week as far as the other languages. If you talk to her another language, she will talk back in that
    02:53:52.360 –> 02:54:03.530
    Amanda Kahlow: language. I can’t promise it’s going to be perfect right now we are signing some of the biggest enterprise Sas companies in the world who have made this a requirement. So it is on our immediate roadmap to support all languages.
    02:54:03.774 –> 02:54:14.289
    Amanda Kahlow: But you’ll have an if you type it, you’ll have a great dialogue back and forth, but try, and you know we’ve had people talk to her in Mandarin, and she does a decent job. I wouldn’t say she does a great job.
    02:54:15.610 –> 02:54:30.660
    Julia Nimchinski: Question from Andrew. What is the AI model? And limitation behind the video? Is the lady only available in a half body speaking head, or you can produce a sort of Tiktok influencers version.
    02:54:31.020 –> 02:54:40.919
    Amanda Kahlow: Yeah, you can do it in anything. So actually, we have. So myself, I this morning, right before I got on this call, I was filming my own video for my own photo reel
    02:54:41.080 –> 02:55:00.570
    Amanda Kahlow: version of myself. So I don’t necessarily recommend that for everyone. But there are certain cases where it does make sense. So you can clone anyone you want. So you know, new relic is a customer of ours. They’re gonna if they do the face, they’re gonna pick the face that they want. We also have the orb, so it doesn’t have to be a face. So if you’re like, Oh, my gosh! The face is overwhelming. We also have avatar.
    02:55:00.710 –> 02:55:09.900
    Amanda Kahlow: I kind of think avatars yesterday’s news. We did have the avatar, you know. We just released the photo rail this week, but we’re moving so fast.
    02:55:10.070 –> 02:55:30.300
    Amanda Kahlow: so yes, you can clone any person or actor, and then, by their likeness, I’m going to follow up with all of my sales calls within my signature, it’ll say, you know. Have a conversation with me. Continue it anytime you want, and it basically spins up a zoom, and you can talk to the AI superhuman Amanda. Ask her anything about my life and about the product, the company.
    02:55:30.450 –> 02:55:46.029
    Amanda Kahlow: and have a full conversation. So yeah, Jacko, from winning by design is about to release his face so, and he’s going to be selling his services with his face and his brand. It’s not for everyone, for sure. I don’t recommend it for everyone, but in you know, in certain cases we’ll say yes.
    02:55:47.330 –> 02:55:56.889
    Julia Nimchinski: Question from Anna. Can she be trained to also learn multiple job positions like ability to become an agent and develop a solution in clay? For example.
    02:55:57.750 –> 02:56:12.570
    Amanda Kahlow: Yeah, absolutely. She can take on any job function across your organization. I like to say, if it has a beginning, middle, and an end. She’s really good at it. So if she has, if it’s a goal like in sales, the goal is to qualify or to sell or to capture information and get the next meeting.
    02:56:13.010 –> 02:56:28.419
    Amanda Kahlow: In customer success. The goal is to onboard. The goal is to upsell the goal is to share new product releases. So whatever that those goals happen to be across your organization. I’m even talking to some of the largest insurance providers where she’s going to be sharing the health benefits and help you pick your
    02:56:28.420 –> 02:56:53.340
    Amanda Kahlow: as you get your plans every year like help. You pick your plan right like. Have a conversation. We built an AI a while ago for an Alzheimer’s caregivers facility. So actually, one’s near and dear to my heart because my mom passed of dementia a while ago, and I just I remember her caregiver saying, I’m doing this for 20 years, and I’ve yet to have a dementia or an Alzheimer’s patient, so she didn’t know what she was dealing with. Right? So as she’s having that conversation.
    02:56:53.340 –> 02:57:11.139
    Amanda Kahlow: so the use cases are endless. I am really trying to stay focused on. Go to market right now. I think that that is my biggest challenge, as the CEO is to not let us go off on a tangent and do all these other things, but it’s super exciting as a business, because there’s a lot more that we can do as we start to scale the the company.
    02:57:12.240 –> 02:57:22.349
    Julia Nimchinski: Question from Nicholas. It sounds like a great idea to feed Mindy. The hard questions in developing the Llms. And effectively dealing with various scenarios?
    02:57:22.620 –> 02:57:31.570
    Julia Nimchinski: Is there careful consideration taken in the types of questions being asked through chunking to ensure good conversational prompts.

  • [ 02:57:30 ] Amanda Kahlow: Yeah, I mean, that gets into how we’re doing it. So it’s a combination of using prompting and using embeddings. And it’s not. It’s a combination of deterministic with generative. It’s not a full, generative approach. Full, generative approach would make her hallucinate more than we would want her to. So she goes down a goal based path with having the generative of how she says it.
    02:57:53.190 –> 02:57:58.199
    Amanda Kahlow: So it’s it’s a bigger. If she were here I’d have her pull up our architecture, slide and and jump in and show you more.
    02:57:59.960 –> 02:58:06.739
    Julia Nimchinski: Question from Claudia. Can you speak to the onboarding process to give her the right data to train her.
    02:58:07.350 –> 02:58:18.200
    Amanda Kahlow: Yeah, the 1st thing we do is we scrape your website so we will. I know when I was in a sales cycle for a very large data engineering company.
    02:58:18.940 –> 02:58:47.229
    Amanda Kahlow: We just scraped their website and we came back. And they said, Oh, my God, this is better than our best sales engineer. But as we onboard real customers we will bring in, we take your best like gong or chorus recording calls. Right? Take your best calls and transcripts and bring those into the conversation, and then we link into your sales. Enablement docs as well, so you can feed us your your pitch, decks, etc, and then we go through a few workshops to help us understand. What does the conversation flow really look like, and what is the outcome that you want her to achieve.
    02:58:48.830 –> 02:58:55.510
    Julia Nimchinski: Thank you so much, Amanda. Flow. We are about to transition to your demo, Amanda. It’s such a treat. It’s always
    02:58:55.510 –> 02:58:55.990
    Julia Nimchinski: thank you
    02:58:55.990 –> 02:59:01.400
    Julia Nimchinski: hosting you on the summit, and it’s the 1st time we actually saw the tech in action. So thank you.
    02:59:01.790 –> 02:59:03.460
    Amanda Kahlow: Thanks for having me. It was super fun.
    02:59:03.830 –> 02:59:05.689
    Amanda Kahlow: Good luck, Flo! Kill it.

Table of contents
Watch. Learn. Practice 1:1
Experience personalized coaching with summit speakers on the HSE marketplace.

    Register now

    To attend our exclusive event, please fill out the details below.