Latest / The Koerner Office - Business Ideas and Deep Dives with Chris Koerner / Here's Exactly How You Can Let Robots Do all the Work for You | Ep. #183
Transcript
- 0:00People don't realize like they think AI agents or this like
- 0:02pipe dream, this thing that's gonna happen at some point in
- 0:05the future, I'm like, no, no, no, it's happening right now.
- 0:07I've been saying that, but here it is.
- 0:08People have success with this, like we've got one in particular
- 0:11starting to generate like $10,000 a month with these ND
- 0:14agency. It doesn't take a lot.
- 0:15It's just go to Lindy dot AI slash templates and we have like
- 0:18hundreds of templates here. I do believe AI agency is of the
- 0:21future because most people these don't know how to set up these
- 0:24AI agents or even if they know they don't have the time to.
- 0:27If someone were to launch an agency, what industry would you
- 0:30attack first? Wow, OK, my brain is buzzing.
- 0:35I just finished an interview with Flo.
- 0:38He's the owner of Lindy. Lindy is basically the AI first
- 0:41Zapier of the future. And I'm a user and a fan and I
- 0:44wanted to find out how can we use Lindy to make money or to
- 0:47save money or to save time. Very granularly, actionably, he
- 0:50shared a screen. My mind is blown that period.
- 0:53I'm not hyping this up. AI agents are here.
- 0:56I thought they were just here in theory and it was just a cool
- 0:58talking point and it was an easy way to raise a bunch of money
- 1:01for something that wasn't ready yet.
- 1:02But they're here. Agents are here, and this was
- 1:04incredible. So you're going to love it.
- 1:05Share with a friend and enjoy. Well, why don't you just give us
- 1:08a high level overview of what Lindy is and what problem it
- 1:11solves? Yeah, we are a no code platform
- 1:14that lets you build AI agents. What it solves is that, well, AI
- 1:18agents work for you. They're like AI employees.
- 1:21So it's, it's just the same sort of problem that like a human
- 1:24employee would solve for you, except it's way cheaper, way
- 1:27faster, it works 24/7. It doesn't go on strike.
- 1:31I know, I'm, I'm French, but we, we haven't implemented the
- 1:34strike until yet. People use AI agents to, I mean,
- 1:37save time, save money, make money.
- 1:39You automate your customer support, you automate your lead
- 1:42generation, your lead outreach and and so forth.
- 1:45Where would you draw the line between an AI automation and an
- 1:48AI agent? How do you differentiate between
- 1:50them personally? An AI agent is powered
- 1:54throughout by an LLM, which is at every point aware, you know
- 2:00the Camela Harris quotes like you know everything that comes
- 2:03before you and the context in which you exist and what that's
- 2:06AI agent versus AI automation. It's like they understand
- 2:09everything that just happens to have this memory and it's both
- 2:13short term and sometimes long term memory.
- 2:14They understand their history and they learn over time.
- 2:18So for example, an AI agent can engage in a conversation with
- 2:22someone, right? Because it understands
- 2:24everything that was just said. And so if you add a message to
- 2:27the conversation, it knows what it means because it's seen as
- 2:29the rest of the conversation. And AI automation is more like
- 2:33each step is an island and if you want some context to bleed
- 2:37over from one step to the next, you need to build that in,
- 2:41right? Not all the context
- 2:42automatically bleeds over. So like an AI automation is
- 2:44really useful for like really simple like ABC, like once and
- 2:48done things like, hey, a new date comes in, you research him,
- 2:51you qualify him, you send him an e-mail.
- 2:53Like that's an AI automation. And AI agent is like for more
- 2:56advanced use cases, any, any use case where you're going to have
- 2:59a conversation with human requires an AI agent.
- 3:03And any use case where you're less than 100% sure of what
- 3:06should happen next at an even point should have an AI agent,
- 3:10because the AI agent decide what to do next.
- 3:13So like an agent needs to be more open-ended right?
- 3:16More variables. It can be more open-ended.
- 3:19Exactly, yes. OK, maybe this is a good
- 3:22example. I I'm running Facebook ads right
- 3:24now for a pickleball club that I'm launching.
- 3:26It's just like a side project and I have a Zapier automation
- 3:30set up to take my Facebook leads and put them in a Google Sheet.
- 3:34Super dead simple. Then people also fill out a type
- 3:37form where I get their phone number and I'd like to call them
- 3:40and get feedback on what they think about this club.
- 3:42An agent might take it a step further, give them a call using
- 3:4611 Labs or Synthlo or any of those other things that you
- 3:48probably integrate with and say, hey, how much would you pay for
- 3:51this when you want to be? What's your shirt size?
- 3:53Because founding members get a T-shirt.
- 3:55Like it would go 10 extra steps for me. 100% and like phone
- 4:00calls is like an excellent example.
- 4:01It's it's one of the top use cases for AI agents and and for
- 4:04us, like we do support phone calls.
- 4:06I know you like to keep it tactical.
- 4:07So like I'm happy to do like a demo later on of of a phone call
- 4:10using Mindy, but it's impressive.
- 4:11It's it's kind of surreal to like it heats different once you
- 4:14hear the voice of the agents and you're able to talk to it on the
- 4:16phone. What would you say like 80% of
- 4:19your users are using Lendy? For what type of tasks?
- 4:22It's like 3 big categories of use cases.
- 4:26Number one is sales. So it's like lead generation,
- 4:30lead enrichment, lead qualification, lead outreach.
- 4:33Like you can have an AI agent like personalize it's lead
- 4:36outreach at scale, like really fast, really affordably #2 is
- 4:39customer support. And because we have like
- 4:41thousands of integrations. And so you can get your AI
- 4:44agents to basically entirely automate your customer support
- 4:47over any channel like SMS, e-mail, in.com, Zendesk, Fresh
- 4:51desk, WhatsApp phone, like you name it, you can, you can do it.
- 4:55And then the the third really big one is, and it's sort of a
- 4:59universal one really. It's like personal assistance.
- 5:02So anything e-mail related or meeting related.
- 5:05So like I have Lindy schedule my meetings for me by e-mail.
- 5:09I have Lindy prepare me for my meetings.
- 5:11I have a phone Lindy that I call.
- 5:12It's like, yo, what's on my calendar today?
- 5:14Can you help me clean it up? I have Lindy recalled my
- 5:16meetings and take notes. An e-mail, you know, like Mindy
- 5:19drafts my e-mail triage is my inbox.
- 5:22She sends me a text messages or gives me a call if I have like a
- 5:24really time sensitive e-mail that they need to take attention
- 5:26to that kind of stuff. So you will get a call from AI
- 5:31saying flow dude check this e-mail.
- 5:34The pipe just exploded in your rental house.
- 5:36We've got to get someone over there.
- 5:37Yeah, hopefully that doesn't happen too often, but yes, like
- 5:39I would say once or twice a week I receive a phone call from a
- 5:42Lindy that's like flow. You really want to take a look
- 5:43at this right now and you know, about once a day or so I receive
- 5:47a text message from Lindy that's like flow.
- 5:49You received an e-mail from Bob asking about this urgent thing.
- 5:52And I can actually e-mail message back and forth with
- 5:54Lindy. I can be like, wow, I'm in the
- 5:55meeting, like tell him I'll be right back.
- 5:57And she just like sends the e-mail.
- 5:59Wow. So she'll move that message over
- 6:01to Bob. Well, it's cooler than just
- 6:03moving it over because it's also just like a copy paste of like I
- 6:05have to like type my e-mail. I can literally just be like,
- 6:08tell him I'll be right back and she's going to draft an e-mail
- 6:10saying that. Wow OK I thought I was using
- 6:15Lindy a lot but I haven't even scratched the surface.
- 6:18I haven't scratched it either. I made the dumb thing and it's
- 6:21like every day it surprises me in in what it can do.
- 6:23Yeah. I think it's interesting that
- 6:26you said sales first because like everything I've noticed
- 6:29about business and all these SAS products is like it needs to be
- 6:33able to tie to generating revenue.
- 6:35Like that is the easiest, lowest hanging fruit use case to
- 6:37anything. Like if I know that this X
- 6:40dollars per month that I spend on this tool can return Y
- 6:43dollars per month, then I'm going to be your first customer.
- 6:46Is that kind of what you're seeing?
- 6:47Yeah, Yeah. I mean, that's obviously like
- 6:49the fastest way to ramp up your revenue is to offer other people
- 6:53to ramp up their revenue. And in a way it's a no brainer
- 6:57all why? Because I don't know about you.
- 6:58Like the first time I heard that's the job of an SDRI was
- 7:02like, why is it a human doing this?
- 7:03This is insane that like we need we need a person doing this.
- 7:06So yeah, all of these jobs are are being augmented to the very
- 7:09least and and automated actually quite rapidly.
- 7:14Now what are like the the the 20% use cases that are really
- 7:18interesting but less common for Lendy?
- 7:21Well, seeing quite a bit of of marketing as well, like people
- 7:25automating to automating content creation using Lindy.
- 7:29I need to coin the term for this.
- 7:30Like basically reports, like AI generated reports, because AI is
- 7:35really good obviously at ingesting a ton of content.
- 7:37Like you could upload, upload like a 200 pages long PDF to
- 7:40collect the GPT and ask for summary.
- 7:42And it's basically reading the 200 pages book in like 2 seconds
- 7:45or something. And so some of my favorite
- 7:48agents, we have an agent that sits at what I call the boundary
- 7:52between the company and the outside of the company.
- 7:55So every customer facing interaction, we have that
- 7:59agencies. So every ticket, every e-mail,
- 8:02every meeting that we have with, with the customer or with the
- 8:05prospect like that agencies. And then at at the end of of the
- 8:09day, it sends a, a message and it actually sends it to the
- 8:12entire company. It sends it on Slack.
- 8:14So it's the general channel on Slack sends a message to be
- 8:17like, hey, this is what's happening today as the boundary
- 8:20of the company. And then you can engage in like
- 8:22an interactive Q&A, like back and forth with it on Slack and
- 8:25be like, oh, wait, there's this like bullet pointing here where
- 8:27you're like, Bob was really unhappy with XYZ.
- 8:30You can tell me more about this and I'll reply to you.
- 8:33I love that. And we're seeing more and more
- 8:34use cases of that. Like, I think people
- 8:36underestimate the extent to which companies are basically
- 8:39black boxes. Like you don't really know
- 8:41everything that's going on in your business.
- 8:42Agents already turning on the light for for a lot of
- 8:45businesses. Yeah.
- 8:46I mean, I could even see it like it, let's say I'm the CFO for a
- 8:50publicly traded company and I see that Slack update the
- 8:52summary come in, what's happening in the boundaries of
- 8:54the company. And it's like, hey, Bob, the the
- 8:57customer, Bob had a complaint about XYZ product.
- 9:00And I see that as an executive. And I say, why is that even
- 9:03launched? We launched that product.
- 9:05Like that was like, they're probably seeing like major
- 9:07things. That's like, holy crap, like 30
- 9:10people in the company knew about this but no one thought to tell
- 9:12me, you know, until this automation.
- 9:15Another thing that I 100% sometimes it flags, Oh, customer
- 9:20X was really frustrated with all sales Rep during the call.
- 9:25And I'm like, oh, and it's like, can you send me the medium
- 9:27recording please? And it's like, yeah, right here.
- 9:30It changes the whole company because again, it's like all of
- 9:32your walls are transparent. Everything's every, everyone
- 9:34sees everything. And as a result, like learnings
- 9:37spread a lot faster as well, right?
- 9:39You can be like, look like this customer got really frustrating
- 9:42at the moment who said this? So let's not say this anymore.
- 9:45You see that it's like customer got very upset at our employee.
- 9:49John, I want to listen to this. So you listen to this and you
- 9:51learn that John just said what our pricing was in a very nice
- 9:55way and they were just an irrational customer.
- 9:57Can you type back to it and say, hey, I don't really want to see
- 10:01stuff like this anymore because that's just an irrational
- 10:03customer. Can you like train it as you go?
- 10:05100% And so your question earlier about like AI agents
- 10:08versus AI automations agents learn continuously.
- 10:11So you could literally just tell it that natural language like,
- 10:13hey, don't tell me about this Robin Cold world and it's going
- 10:16to remember it. It's going to like keep it in
- 10:18mind moving forward. I do that with I have like an
- 10:20e-mail triage agent that just triages my inbox everyday and I
- 10:25just have like 3 buckets. I have urgent, I have FYI and I
- 10:28have archive. And when I see an e-mail make it
- 10:31in the wrong section, I just like forward it to my e-mail
- 10:34triage agent. Like my e-mail triage agent has
- 10:37its own e-mail address and so I just 4 wheeled these emails.
- 10:41I'm like, hey, this e-mail does not belong in an alternate.
- 10:42Like this kind of e-mail belongs in in FYI and it can just mean
- 10:46all these horror rules. It could be like, hey, emails
- 10:48about this topic or emails about recruiting or whatever.
- 10:51Just just do this moving 4 wheeled.
- 10:52Now are you still at the mercy of whatever LLM you're using on
- 10:56the back end? Because like cha GPT, it's like
- 10:58stop using M dashes, seriously. Stop using emojis, seriously.
- 11:02And it doesn't, right? So are you still at the mercy of
- 11:05that? You are at the end of the day it
- 11:07it is powered by an LLM. I will say though, are you using
- 11:10GPT 4-O? 4 O and three O It depends.
- 11:13I always recommend people not to use four O.
- 11:15It's it's a really bad model, frankly.
- 11:17Like it's fine for like day-to-day conversation, like to
- 11:20power workflows, it's not good. We default to cloud full sonnet,
- 11:25which is excellent. And we also offer cloud 3.5
- 11:27sonnet, which is one of my favorite models.
- 11:29So yes, at the end of the day it's 4 wheeled to buy an LLM but
- 11:32like LLMS are getting really good.
- 11:34Yeah, I said three. I meant O3 and I'm using Cloud
- 11:38Sonnet 4 point O for writing a lot and it is the best.
- 11:41It's really, really. Good.
- 11:42It's so good. With a great context window.
- 11:44Speaking of context windows, is Lindy also at the at the mercy
- 11:48of the LL Ms. context window? Like technically how does that
- 11:51work? Yes, every agent is just an LLM
- 11:54at the bottom right. So yes, at the end of the day,
- 11:57we all subject to the same context window limits as as the
- 11:59LLM offers. I will say though, like modem
- 12:02context windows are, are huge. Like they go up to like 1 or 2
- 12:05million tokens. That's like I see what's the
- 12:07point of like 1000 pages book. It's it's very rare that we hit
- 12:11context window limits anymore. And even when we hit them, what
- 12:14we do is that we just compress the context window.
- 12:17We jam as. Much more efficient.
- 12:19That's right. And it works really well.
- 12:20And like you, basically context windows are for all intents and
- 12:23purposes, they're basically getting infinite.
- 12:25It's been a few months since I've looked for this solution
- 12:27but I I just want something to read all of my emails like
- 12:32through the past and to be able to just learn so I could train.
- 12:37I could pass it on to an employee for customer support
- 12:39request. Is there a clean way of doing
- 12:41that today, or are there too many privacy concerns that
- 12:43prevent that? Is there is a way to sort of
- 12:46hack it together, for example, so like the way you would hack
- 12:50it on Lindy is you would expose your inbox in Gmail.
- 12:54Does like an option to export every e-mail you've sent or
- 12:57received and probably sent is the most important one.
- 13:01And then you would upload that to Lindy as a knowledge base and
- 13:04then you would ask Lindy to use that information to draft your
- 13:07emails or respond moving forward for you or what not.
- 13:11OK. But like me, I've got a polling
- 13:14company and we have 8 different businesses and they all, a lot
- 13:17of stuff just funnels to my main inbox.
- 13:19What how would you be able to sort that out?
- 13:21Like if this business then respond this way, you know,
- 13:24whereas if, if I was like the customer support lead for one
- 13:27business, it'd be very straightforward.
- 13:28It's like, here's how I answer all the random questions, you
- 13:31know? Yeah.
- 13:32So you can set up branching in agents, including in Mindy.
- 13:34So you could just have like have a condition like a branching
- 13:37node is what we call it. And then you can be like, hey,
- 13:40if this business go here, if that business go there, and then
- 13:43each of these branches would lead to a different knowledge
- 13:45base. Can you use Lindy to to shrink
- 13:48the context window down? Because that's something that I
- 13:50might actually butt up against the context window.
- 13:52How would that work? Well, you wouldn't butt up
- 13:54against the context window for this use case because the point
- 13:57of the knowledge base is that it's got, it's got a huge corpus
- 14:01of knowledge, sometimes like trillions of tokens, but then
- 14:04it's using RAG like a retrieval augmented generation.
- 14:07So it's basically searching. It's like the equivalent of a
- 14:09Google search for this huge corpus of knowledge.
- 14:12So then it actually context, then it injects into the LLM and
- 14:16the agents context window is actually quite limited.
- 14:18It's just a couple thousand tokens.
- 14:19So you you probably wouldn't run into the limit.
- 14:22That said, if you do run into the limit, yes, we also have a
- 14:25step that you can inject in your agents.
- 14:27That's like when you reach this point comprise the context
- 14:31window and you can pass it to prompt.
- 14:33That's like compress the the context window and before you do
- 14:36that extract XY and Z information.
- 14:39Gotcha. OK, cool.
- 14:41Has there ever been an automation that it was just kind
- 14:44of like a this is so crazy that it worked?
- 14:46Like I can't believe this actually worked.
- 14:48And what was that moment or automation like?
- 14:50Yes, I would say there's two categories of automations, but
- 14:54we'll like I can't believe this works.
- 14:55There's one where we'll like. I can't believe this works just
- 14:58because the user set it up in such a crazy way.
- 15:02An inefficient way. Yeah, you could call it that.
- 15:05It's like very often they just have like 1 giant AI agent step
- 15:07with like a huge prompt to like 50 actions in that prompt.
- 15:11It's like Oh my God. And it works.
- 15:12And the prompt doesn't even really make sense when I really
- 15:15as a human, I can barely understand what the person is
- 15:16saying and somehow the agent is figuring it out.
- 15:19I'm like, how does this work again?
- 15:20I think that's the strength of the AI agents is that they are
- 15:23really forgiving like automations and coding in
- 15:26general. Like you just get one little
- 15:28thing wrong and the whole thing collapses and stops working.
- 15:31AI agents, the limp log like you could, you can remove
- 15:34characters, you can remove entire lines.
- 15:36Like the AI agent is always going to try to figure it out.
- 15:39And then there's the second category where we'll like, we
- 15:42can't believe it works just because the use case is so
- 15:44complex. It's not as much like the user
- 15:46screwed up, like the user did a good job.
- 15:48It's just like, Oh my God, this is an insane use case.
- 15:50Like I think some of my favorite AI agents are like sales and
- 15:54like SDOAI agents. So we have this one that engages
- 15:58with a lead on a web chat interface and it collects
- 16:02information. It qualifies the lead.
- 16:04And then if the lead wants a meeting, the agent books the
- 16:07meeting. So it looks at the calendars of
- 16:08of the company. It like books the meeting.
- 16:11It sends a confirmation e-mail, it sends a confirmation text.
- 16:13It calls the person on the phone before the meeting to confirm
- 16:16some information after the meeting.
- 16:19So it it records the meeting, it joins the meeting as well as as
- 16:22the sales Rep text notes. And after the meeting it sends a
- 16:24follow up to be like, hey, thank you for your time.
- 16:26This is what we discussed. And the whole time it keeps in
- 16:30context all the previous interactions that it had over
- 16:33all these modalities. So like the web chats, the SMS,
- 16:36the phone call, all that stuff. So like if it goes on the phone
- 16:38and it's emailed back and forth with you before, it will mention
- 16:41that on the phone. I'll be like, hey, Bob, like
- 16:42we've just emailed. We talked about this like how's
- 16:45your day going? So seeing exactly just like
- 16:48holy, like that's when I was like, it's getting real.
- 16:50Like it's, it's happening, you know?
- 16:52Oh man, in that case is it telling the user I'm AI?
- 16:57You can't basically it's not at the point yet.
- 16:59Well, like you would be confused by it, especially when you have
- 17:01it on the phone. You know, it's an AI.
- 17:03The way that users set it up is like, yes, it is obvious that
- 17:05it's an AI also because like on the web chat, it just replies
- 17:08too fast and all of that stuff. We have some customers pretty
- 17:11frequently, actually. They ask us, can we make the
- 17:13agent slower? It's like too fast.
- 17:15Yeah. That's what I would want as a
- 17:17human. Yeah, yeah, yeah, yeah, yeah.
- 17:19In this in this case you can tell it's in the eye.
- 17:21I mean, I've been setting up like e-mail automations for a
- 17:23while where like if someone emails me this with these
- 17:26keywords respond with that. And very few platforms allow me
- 17:30to put a pause in there. Kind of like Zapier, you can put
- 17:32a pause in between something. And I think that's a big part of
- 17:35it, because if, if I send someone a critical e-mail and I
- 17:38get back 4 paragraphs within 10 seconds, I'm going to cock my
- 17:42head as I look at that, you know?
- 17:43Yeah, OK, I'm going to discount this e-mail.
- 17:45Whereas if it takes 1/2 hour to get it, but that's all.
- 17:48I'm sure that's like super easy with Lindy.
- 17:51Yeah, 100%, yeah. Just like Zpio, you can
- 17:53introduce like a waiting step. Yeah, Yeah.
- 17:56I'll give you an example of something that happened actually
- 17:58just a couple of weeks ago. I have Lindy's schedule, main
- 18:01meetings. So if someone sends me an e-mail
- 18:02and it's like flow, let's meet. And I do want to meet ICC, my
- 18:05Lindy assistant to my e-mail like, oh, sounds good.
- 18:09Lindy will help us find things, but then she goes back and
- 18:11forth. She looks at my calendar, She
- 18:13agrees all the time with the other party.
- 18:14She sends the calendar invite. And then a couple of weeks ago,
- 18:17there is this person I was supposed to meet with on that
- 18:19day, actually. And last minute she sends an
- 18:22e-mail in the thread and she's like, Flo, I'm sorry, I have to
- 18:24consult to this meeting. My son just had an accident.
- 18:26I have to go to the hospital and Lindy sees the e-mail.
- 18:30She's on the e-mail thread and she's like, I'm so sorry to hear
- 18:32about your son. Which by the way, like again,
- 18:34that speaks to your point about the AI agent because she's seen
- 18:37the entire conversation and automation here would have sort
- 18:40of work. They would have been done.
- 18:42It's like, oh, OK, branch reschedule request.
- 18:44It's like here also more times. So like here, it's like, I'm so
- 18:48sorry to hear about your son. And she did not even offer times
- 18:51immediately. So like, take your time.
- 18:53I'll circle back to you later, you know?
- 18:55And then she waited and she woke up like I think 5 days later or
- 18:59something. It's just like, hey, how's your
- 19:00son doing? Is everything OK?
- 19:01Is now a good time to put some time on the books again?
- 19:03Oh my gosh, I know I and that's. Worked that in like I like.
- 19:08Did you set it up to do that? No.
- 19:10And that's what I mean by like, I built the platform, but I'm
- 19:14surprised every day by what it does.
- 19:16I can't even talk right now. I'm just thinking like I can set
- 19:20up an app that does that and it's like my son just got hit by
- 19:23a car. I have to transfer or I have to
- 19:25push and it's like what date works for you?
- 19:28That just. Looks so bad that just I cringe
- 19:31even talking about the hypothetical.
- 19:33And this happens every day, like just the other day, someone else
- 19:36I did his human assistant to the thread to schedule time and the
- 19:40e-mail assistant was like ha ha, you're adding your AI assistant
- 19:42like am I getting replaced anytime soon?
- 19:44And Lindy replied she was like, ha ha, like I don't think I'm
- 19:48replacing anyone anytime soon :) I was like, that's just, that's
- 19:52like, smooth and cool, you know? Right, right.
- 19:56So with like the, the SDR, the sales example, are there
- 20:00companies like it just seems like we're still so early that a
- 20:03lot of the sales stuff is like a party trick.
- 20:05Like there's a dozen companies in Silicon Valley that have
- 20:08raised 10s of millions of dollars for sales agents, right?
- 20:12That's it. Do you have customers that are
- 20:14like wholly relying on sales agents to actually bring money
- 20:18in the door, like end to end? Yeah, absolutely.
- 20:21Well, not end to end like you still need like a human to close
- 20:24them, that's for sure. For now full now.
- 20:27Yeah, I think it'll take a while to replace that.
- 20:31Be it only because humans don't build relationships.
- 20:34I think sales is truly about relationships.
- 20:37Humans have like an anti AI bias.
- 20:40Humans don't build relationships with AIS, they build
- 20:43relationships with humans. So I think like that reason
- 20:46alone, I think humans are going to be in the cells loop for a
- 20:48very long time unless we reach the point where the AI is
- 20:52indistinguishable from the human even on a zoom call.
- 20:55And we reach a point where people are comfortable
- 20:58misleading the customers and not telling them it's any high.
- 21:00But I think we'll pretty far away from that.
- 21:01And I think STPP be messed up to do this.
- 21:04But yeah, yeah, we do have some customers who basically like all
- 21:07they have is a ISDOS, like they don't have human SDRS anymore.
- 21:10Yeah. The example that you gave a few
- 21:13minutes ago of the SDR agent that that does everything and it
- 21:16follows up is the call that it's booking.
- 21:18That's still with an actual human then, right?
- 21:20But the AI agent will follow up what will remind them it will do
- 21:23everything else. That's right.
- 21:25Including sending a prep e-mail to the human right, they'll be
- 21:29like, hey, you have a meeting with Bob.
- 21:31This is what we discussed previously.
- 21:32This is what you need to know do.
- 21:34You guys have any integrations with ringless voicemail services
- 21:38is what that is? No, I don't know.
- 21:41Ringless voicemail drops where there's a company that makes
- 21:43sense. Like slide broadcast does it
- 21:46because like if I'm an SDR agent or if I, if I'm a sales manager,
- 21:50I would want to clone my sales agent's voice with 11 labs and
- 21:54then send the like a, a summary of the call.
- 21:57But I don't want to actually answer the human.
- 21:59And so I would use something like slide broadcast to just
- 22:02drop a voicemail with the sales agent's voice.
- 22:06So they, they think it's him and just say, hey, thanks for the
- 22:09call. Just want to follow up.
- 22:09These are my action items. These are yours.
- 22:11Hope you have a great weekend. And they're just going to assume
- 22:13they they missed the call. Voicemail.
- 22:16Ringless voicemail drops. Ringless voicemail.
- 22:19I'll, I'll look it up. That should be something free
- 22:21choice for wheel to integrate with, yeah.
- 22:23So you obviously need a person to like collect the wire info
- 22:27and hop on the sales calls and warm them up and have the
- 22:30relationship. What types of industries do you
- 22:33think that using sales agents it works the best in today?
- 22:38Because like if I'm selling like copper pipe fittings B to B,
- 22:42probably not great, but if I'm selling like enterprise SAS
- 22:46could be great. What are you seeing works really
- 22:48well for sales? Specifically what industries or
- 22:51price points? Yeah, I had this the same prior
- 22:54as used like pipe fitting another fit Enterprise SAS a fit
- 22:57and I actually think it's the opposite.
- 22:58I think pipe fitting is a great fit.
- 23:00And the reason for that is because the people you sell pipe
- 23:02fitting too or not really tech savvy, like they don't want to
- 23:06jump on a zoom call. They're sort of like on the goal
- 23:08all day. So like if you, if your prospect
- 23:10is on the goal all day, you should, you should probably use
- 23:13one of these voice agents. If your prospect does not like
- 23:16is not tech savvy, you should probably use a voice agent.
- 23:18We actually even have people deploying voice agents for like
- 23:21their own reps who are like field reps like so like they're,
- 23:24they're on the field way. They're like in the car and
- 23:27they've deployed AI agents to help these reps updates the CRM
- 23:31on the go just by making a phone call.
- 23:33So anytime your your prospect is on the go or anytime if you're
- 23:38very busy and you can't pick up the phone.
- 23:39So we actually have like some blue collar workers like
- 23:42plumbers. We have some restaurants as well
- 23:44because restaurants, they're busy at Piga oils, you know, you
- 23:47see them, they're like run around and stuff.
- 23:49And that's also the time when you receive the most phone
- 23:51calls. So we actually have some
- 23:53restaurants using us as like an AI photo receptionist because
- 23:56it's always the same questions like are you open tonight?
- 23:58Are you solving? Yeah.
- 23:59And the photo receptionist comes like make reservations and so
- 24:02forth. And then like the Plumb bill use
- 24:04case is different. It's more like this guy is
- 24:06plumbing. It's not, it's not like like
- 24:08picking up the phone and every phone call he's missing, his
- 24:11business is missing. So I would say those are like
- 24:13the biggest cases where, well, like AI voice agents are really
- 24:17effective. Yeah, I think there are some use
- 24:18cases where there's you almost do more harm than good
- 24:21disclosing that it's AI. Like if I'm calling a restaurant
- 24:24to find out, do you have any tables right now?
- 24:26Can I make a reservation a week from now?
- 24:28Are you open, you know, can I get a gluten free pizza like as
- 24:32a customer? I'm cool with that being very
- 24:34transactional. I would rather talk to a robot,
- 24:37whether I know it's one or not, then like sit on hold or leave a
- 24:40voicemail and I know I'm never going to get called back.
- 24:42Whereas in like the SDR example, something higher ticket or more
- 24:45personalized because I'm not building a relationship with the
- 24:48Hostess at a restaurant. I probably want to say as I
- 24:50answer the phone like, hey, I'm an AI agent, but I can, I'm
- 24:53actually a lot more capable. How can I help you?
- 24:55Yeah. And, you know, some people, at
- 24:57least at first, they dislike talking to an AI agent.
- 25:00And I think a lot of that is because they've been conditioned
- 25:03by, like, the phone trees, right?
- 25:05It's like Type 1 to continue Type 2.
- 25:07And that experience is just so awful that people have just
- 25:10learned, like I, I want to talk to a human people by little.
- 25:12They learn that like this is a fundamentally different breed of
- 25:15product, like a food tree of like your parents were using,
- 25:19you know, and then they actually also end up appreciating it
- 25:22because they, they, they appreciate that like, yeah,
- 25:24you'll never put on hold. You can talk to me if you can
- 25:27stop and you can call it as much as you want.
- 25:29You're not afraid of, of this building it.
- 25:30Like I have my Lindy. I, I have a human assistant as
- 25:33well, but I, I call my Lindy on the phone all the time, actually
- 25:36call, call it more often than I would my human assistant.
- 25:40But even though the human assistant like works for me,
- 25:42like I, I wouldn't feel comfortable calling him like 20
- 25:44times a day and like 11:00 PM to ask for like random stuff.
- 25:47Yeah. I mean as a consumer, I am more
- 25:50than ready for companies like Verizon or these mega companies
- 25:54to just completely get rid of their call centers and replace
- 25:56it with AI voice agents. We just switched to Verizon, our
- 26:00whole family plan and my son's phone for whatever reason, the
- 26:02number would not port over and we called support four times.
- 26:06We went into the store and they ended up putting me on support
- 26:09while I'm in the store and it never got fixed.
- 26:11We ended up having to get a new number and I was just thinking
- 26:14the whole time like I am so ready to talk to a robot here
- 26:17because I wouldn't have to wait. It knows all the long because I
- 26:20know I know our problem was like a very long tail obscure issue
- 26:23and it would know that whereas the Rep doesn't.
- 26:25They have like 4 issues that they work through and so I kept
- 26:28having to be put on hold and transferred and it's the IT
- 26:30can't get any worse. Like this can only be an
- 26:33upgrade, you know so. You can actually also use AI
- 26:36agents for this. I I use Lindy for this online.
- 26:39Like I had also an issue with my Internet service provider a
- 26:42couple of weeks ago and I just sent my agent to do it.
- 26:45She spent half an hour on the phone with these people because
- 26:48Google Fiber, I hate them and half an hour on the phone with
- 26:52them and, and in the entry resolve station.
- 26:55That's that's 1/2 hour back. In my weekend.
- 26:57Wow. Oh man, it's like the robots
- 27:01talking to people who act like robots, but the real robots are
- 27:05actually smart. Yes, yes.
- 27:07I would love for you to show like some interesting use cases
- 27:11on your screen that could apply to like a large amount of
- 27:13people. Yeah, totally.
- 27:15Let me show my screen here. And I'd love to hear about any
- 27:19agencies that are building agencies completely on top of of
- 27:23Lindy, maybe it's a lead Gen. agency or AI automation agency,
- 27:27etcetera. Yeah, I'd love, I'd love to talk
- 27:29about this as well for sure. Let me show my screen right now.
- 27:33This is my chief of staff in this.
- 27:34It's sort of like the the main Lindy I use all the time.
- 27:37And so you can see like just on Sunday, I called, I called her
- 27:39four times. I'm going to call her right now
- 27:42my Lindy number and I put her on speaker.
- 27:47Hi, Flo. How can I help?
- 27:48Hey, Lindy, what's on my calendar today?
- 27:50You have an interview with entrepreneurship and
- 27:52opportunities, 2 meetings with the marketing team, and three
- 27:56interviews with candidates. Yeah.
- 27:58Anything else? Hi, Flo, how can I help you?
- 28:00That's all. I think the demo effect, I think
- 28:03it just keeps saying that. But right here.
- 28:06No, it just keeps repeating. Hi, Flo, how can I help?
- 28:07No, that's all I said you're going to do for today.
- 28:09Jeez Louise, I don't know why I'm not doing that.
- 28:12I'm using the ChatGPT voice feature a lot, Like multiple
- 28:15times a day. It doesn't even hold a candle to
- 28:17this because it doesn't know what's on my calendar, you know?
- 28:21Yeah, I use it for that all day. I use it for like this is the
- 28:25thing I mentioned. Well, like anytime I receive a
- 28:27time sensitive e-mail, it sends me a text message and I'll show
- 28:30you what it looks like in the back end.
- 28:31It's actually really, really simple Floyd there.
- 28:34It's like if I receive an e-mail, you look at my calendar
- 28:37first because you need to know what's on my calendar in order
- 28:40to, to determine if it's a time sensitive e-mail.
- 28:42And then there is this condition, and this is what I
- 28:45mean by AI agents can see everything that just happened
- 28:48before. Like here, I can write a prompt
- 28:50for my condition and the prompt is just like, if this is a time
- 28:53sensitive e-mail from someone asking to reschedule a meeting,
- 28:55I have to date with them. So again, I don't need to inject
- 28:59the context. I don't need to do any of that
- 29:01stuff. It's like an AI agent that
- 29:02received the e-mail that looked at my calendar and now it knows
- 29:05what I'm talking about. And if this is a time sensitive
- 29:08e-mail, you send me a text message.
- 29:10So it's like you received an e-mail message from name about
- 29:13very, very short summary. And here I can just just like
- 29:16add this kind of and here I'm just prompting.
- 29:18And you can see I'm prompting in this case cloud full Sunnet,
- 29:21which again is one of my favorite models ever.
- 29:23Just awesome. What do you love about it?
- 29:25The vibes or Immaculate in the the balance of the youth.
- 29:29It's it's in the models there, you know, it's it's smart, it's
- 29:33cheap, it's like it's affordable, it's fast.
- 29:35I really like it's writing style.
- 29:37Like I really usually cringe when they see LLM writing
- 29:39styles. It's like you can tell it's an
- 29:41AI model here. Like you can barely tell
- 29:43anymore. Yeah.
- 29:45OK, that's amazing. It actually also sends a message
- 29:48to my human assistant that like sends the same text again and
- 29:51again. This is the difference between
- 29:53an agent and an automation. Like here I can be like sends
- 29:56the same text again because it knows the text message it just
- 29:58sent because of that memory. But this time say flow received
- 30:01not your received. I use Lindy all the time for
- 30:04like, well meeting recording. I use Lindy to manage my
- 30:08personal CRM. Every Friday, for example, she
- 30:10wakes up, she looks up my calendar, she looks at my CRM,
- 30:14and then she recommends people to add to my CRM.
- 30:17She's like, you met with these new people that you don't have
- 30:20in your CRM. I checked you don't have them in
- 30:22your in your CRM yet. Do you want to add them to your
- 30:24CRM? And they can just be like, yeah,
- 30:26add AB and C and tag them like C is an investor, B is a sales
- 30:29leader. And then every so often I go to
- 30:32Lindy and I'm like, hey, who? Oh, for example, like we just
- 30:36made like a big announcement a couple of of weeks ago and I was
- 30:39like, hey, who are the AI influence fields that I know?
- 30:43And it just goes in my serums like, hey, these are the
- 30:45influences. Yeah.
- 30:48I use it for my daily digest, so it's basically I'll just show
- 30:51you the flow detail for this one.
- 30:52Every morning it wakes up, it looks at my calendar and then
- 30:55for every meeting and for every attendee of every meeting, it
- 30:59looks for the LinkedIn. It looks for my e-mail history
- 31:02with this person. It looks for my meeting note
- 31:04that are brought together by a Lindy and then it brings
- 31:08together like one nice e-mail with like this is people you're
- 31:11meeting with today. This is the last thing you
- 31:13discussed during your last meeting.
- 31:14This is the things you discussed over e-mail with them and so
- 31:17forth. I'll give one last example and
- 31:18then I'll shut up because otherwise I never stop.
- 31:20No, this is good. I want to learn about what
- 31:23you're doing with deep research as well.
- 31:25Yeah, I have this deep research agent here.
- 31:27Yeah, well, why don't we start with that actually.
- 31:29So deep research, I use deep research for.
- 31:33I really like learning about companies and about founders.
- 31:36And I find that there's actually a lot of knowledge in podcasts,
- 31:41I actually think. This and they're not very
- 31:42searchable. It is not very searchable.
- 31:44NTL agents. Yeah.
- 31:46And so I'll show you an example of how it works here.
- 31:49So here, for example, I wanted to learn about like Mark Benioff
- 31:52or like Zeb Evans. So Zeb Evans is the founder of
- 31:55Click App. And so I'm like, I'll just give
- 31:57you the name, like Zeb Evans, OK.
- 31:59And here it is searching on Google interviews with Zeb
- 32:02Evans, and it's finding 10 results.
- 32:05And then it's going through all these results and
- 32:08simultaneously, we call this an agent's wall.
- 32:10It's really good. So simultaneously, it basically
- 32:12spins up one copy of itself, OK? And it transcribes the video and
- 32:17then it summarizes the video and then all.
- 32:20These summaries one video at a time.
- 32:23Well, no, it does all of it at the same time.
- 32:26OK. And then once the summaries are
- 32:27done, it exits the loop and then it it gives me the summary of
- 32:30summaries. So it's like this is everything
- 32:34that Zeb Evans mentioned. And I can go and ask follow up
- 32:36questions like, ah, tell me more about XY.
- 32:38It's it's an agent, so I can just I can just chat with you.
- 32:40Yeah. Dude, you know what would be
- 32:41cool is that same thing, except let's say Zeb Evans has 10
- 32:45interviews, 10 period total, and in each interview there's like
- 32:4960 to 80% of overlap. He's kind of saying the same
- 32:52thing. I would love an agent that shows
- 32:54me all the non overlap. That's what it does.
- 32:57That's what it does. Like the childhood stories and.
- 33:00Yeah, yeah, yeah. So you can see here, for
- 33:02example, like key themes across interviews, right?
- 33:06So like, this is an agent that has seen the summaries of all
- 33:09the YouTube videos and it's giving me its own analysis.
- 33:15My gosh. So I can ask actually like here,
- 33:17like, what are some of the things that he says across the
- 33:22most interviews? I mean, for my podcast prep,
- 33:25this would be amazing. I could have done this with you.
- 33:28You could have. You should have.
- 33:30Looking across the interviews, here are all the themes that he
- 33:32consistently repeats. 1% better over a day.
- 33:34Near death experience is 6 total.
- 33:38Yeah, yeah, he like I get like quotes on interviews and so
- 33:42forth. Recruiting.
- 33:44I'll show you this one cuz that's that's one of my
- 33:45favorite. So I spend a lot of my time
- 33:47recording like I always joke that a founder is basically just
- 33:50like a glorified recruiter. We all right now hiring, by the
- 33:53way, we all hiring. Hit me up.
- 33:56OK. You heard it all, hiring across
- 33:57every function, but one of the ones that we're hiring for is
- 34:01sales leaders. So here I have this linear
- 34:04recruiter and if you look at the timestamps, look, it's like full
- 34:0713 PM on Tuesday, right? And I'm like, hey, find me
- 34:10directors of SMB sales from one of these companies and I give
- 34:13you the list of companies. And she's like, all right, let
- 34:15me look. She's searching and she's like,
- 34:17I found one candidate and then she says let me try another
- 34:21search. And again, it's an EI agent.
- 34:23I didn't need to ask it to do this.
- 34:26It's like smart enough to know. Like, it's probably, he's
- 34:27probably not going for one candidate, you know?
- 34:29So like, let me let me search for a rule.
- 34:31It's like, I found, I found these guys here and I'm like,
- 34:34yeah, like, keep searching. So keep searching.
- 34:37And then it finds, it finds these people.
- 34:39It's like, OK, like, tell me which ones of those you want me
- 34:41to reach out to. And I'm like, yeah, all of them,
- 34:43all of them look good. And it just just reaches out.
- 34:46And so here again, it's it's kicking off an agent's womb
- 34:48where it's like for each person, I'll just click in the in the
- 34:52subtasks here for each person. First it checks, it searches my
- 34:56inbox to check if I've already reached out before because
- 34:58sometimes I have. So it's like, OK, you've not
- 35:00reached out to this person before since you haven't reached
- 35:02outlet me draft an e-mail for them and boom, hey, this is Flo.
- 35:06You know, we're looking for a head of sales.
- 35:07Would you be interested in your conversation and all of this?
- 35:11I wonder if there's like another time step I can look at.
- 35:13Well, yeah, I can because like, this was kicked off at full 13.
- 35:15And this, yeah, at full 18 the outreach started.
- 35:20So in 5 minutes it founds the people I was looking for and it
- 35:23sends the e-mail and then you can see it sends follow up.
- 35:25So like it woke up again two days later and then again two
- 35:28days later and just like kept sending follow up emails.
- 35:31Oh my gosh, I can't even like think of my next question
- 35:35because I just want to log off and just start building these.
- 35:38For some reason, it's like it's real, like people don't realize,
- 35:41like they think AI agents saw this, like pipe dream, this
- 35:44thing that's gonna happen at some point in the future.
- 35:45I'm like, no, no, no, it's happening right now.
- 35:48I've been saying that like I tweeted a couple months ago,
- 35:51like everyone's talking about AI agents, who is actually using
- 35:53them today in their business and how and like I just heard a
- 35:57bunch of like third hand stories of, but here it is.
- 36:01How can the general public like I know you have like a
- 36:03repository of you call them flows.
- 36:06I we call them Lindy's, but we're going to our name agents.
- 36:09And how can people just copy and paste this with their own use
- 36:11case? Yeah, you just go to Lindy dot
- 36:13AI slash templates and we have like hundreds of templates here.
- 36:16I think we have one for the recruiter.
- 36:17Actually. It just showed lead generator.
- 36:20We should have one for recruiter.
- 36:21I think it's just missing from here.
- 36:23OK, I would like to build 1 where I plug in like 20 YouTube
- 36:26channels of other channels in my niche and it sorts them from
- 36:31most to least viewed videos. It finds the person that was
- 36:34interviewed and then it finds them on LinkedIn and starts
- 36:37messaging them. Hey, you were on this podcast.
- 36:39I have a very similar podcast. Do you want to come on mine?
- 36:41Like, that's pretty simple, right?
- 36:43Yeah, I mean those like a scrape.
- 36:45YouTube via Lindy chat here. Yeah, yeah, absolutely.
- 36:50Oh man. What about building agencies on
- 36:52top of Lindy? What are you seeing there?
- 36:55We all seeing we're starting to see people have success with
- 36:57this. Like we've got one in
- 36:58particular. I forgot the exact number that
- 37:00he told me, but I think it's starting to generate like
- 37:02$10,000 a month with his Lindy agency.
- 37:05So what it consists of is like he is posting about us on on
- 37:09LinkedIn and he's making videos about us on his YouTube channel
- 37:12and he's like, hey, check out this like cool Lindy.
- 37:15He's basically doing what I just did, except he's selling his
- 37:17services, right? So it's like, hey, like check
- 37:19out this cool Lindy. This is how I've set it up.
- 37:21This is how it works and it's a cool demo too.
- 37:23Like this demos or just compelling.
- 37:26And then she, you know, he gives the template in the, in the
- 37:28description and, and, and he's got a lot of people reaching out
- 37:31to him who are like, I don't want the template, I want you to
- 37:33sit there for me. And he also feels that's
- 37:35obvious. And then he's, I think he
- 37:36charges like a one time fee. I'm not familiar with his
- 37:39pricing but I believe he's charging like a one time setup
- 37:41fee and then like an ongoing maintenance fee of like 500 or
- 37:45$1000 a month which is nothing. Like if you think of AISDR it's
- 37:50just like pays for itself and like 3 or something.
- 37:53And then he's he's mostly finding customers just by
- 37:55putting out free content. That's right.
- 37:57That's right. Yeah, and most of them will go
- 37:58on to either do nothing or implement it themselves, but a
- 38:01minority of them will say, hey, just do this for me.
- 38:03That's right. And it doesn't take a lot, you
- 38:04know, if you're talking like 500 bucks a month for custom meal,
- 38:06like 10 custom meals, you've got $5000 a month.
- 38:08Like I do believe AI agencies are the future because most
- 38:12people, these don't know how to set up these AI agents or even
- 38:15if they know they don't have the time to.
- 38:17So AI like address address both of these issues.
- 38:21If someone were to launch an agency, what industry would you
- 38:24attack first? I would only say any industry
- 38:28that you're familiar with, any industry where you've got an
- 38:31audience and you understand the industry and the challenges it's
- 38:34facing, just focus on that because every industry can
- 38:37benefit from AI agents. Again, like AI agents are
- 38:39basically AI employees. Every industry has got
- 38:42employees, you know, so every industry could benefit from
- 38:44having AI employees instead because they're just better to
- 38:47build a faster and more intelligent.
- 38:49So yeah, just any industry that you're familiar with.
- 38:52What about using Lindy for lead generation, either for your own
- 38:55business or in selling leads as a business?
- 38:59Are you seeing any of that? Yeah, absolutely.
- 39:00So look, the recruiting use cases just showed that's kind of
- 39:05like a yeah, yeah in use. Case like, like we're creating
- 39:07expensive leads too. That's the 100%.
- 39:10So and it's really targeting like it's really finding exactly
- 39:12what I asked for here, right? I was like, hey, like directors
- 39:15of sales at like like London here, like Dropbox and so forth.
- 39:19So it's yes, we've got a bunch of lead sources integrated in
- 39:24Lindy. The best one is, is one that
- 39:26called People Data Labs. It's just like a lead search.
- 39:28It's almost like Mercury, but not Mercury.
- 39:30Sorry, Apollo. Yeah, better than the Polo.
- 39:33In my experience, the leads are better.
- 39:35It's like less stale otherwise. We also have like a ton of
- 39:38script builds. Like we have like thousands and
- 39:40thousands of web script builds. Like I'll give you an example.
- 39:42We have a customer. He has like a, it's funny, he's
- 39:44got a music studio for video games.
- 39:48Like, he makes like music tracks for video games.
- 39:50And so he set up a Lindy to scrape the Steam store for him.
- 39:54So it gets kind of niche, you know, it's really like.
- 39:57Scrape the what? Is it the Steam web store?
- 40:00Have you heard of Steam? Yeah, yeah, yeah, OK Steam.
- 40:02Yeah, like the the Game marketplace.
- 40:04That's right, that's right, that's right, that's right.
- 40:05So, OK, scrape Steam, find independent studios, like
- 40:09smaller than X, like less than X in sales, because if they're too
- 40:13big, they've got their own in house music person.
- 40:15And then for each of these people, look on the Internet for
- 40:18their contact information and reach out to them.
- 40:21Yeah. OK, this is maybe a difficult
- 40:25question, but if you were forced to nuke all of your partners,
- 40:31like all of your integration partners except 3, which three
- 40:35would you keep? Like which are the most just the
- 40:38most powerful? Doesn't need to be the most
- 40:40common like Gmail or Sheets, just the most absolute powerful
- 40:44integration paired with Lindy. Yeah, I think this one, like the
- 40:48people that at ABS one is, is super awful, like a disability
- 40:51to just find people. I find it is so awesome.
- 40:54I think it's like scraper, like we integrate with the Aptify.
- 40:57Like kudos to the team. I think they've done an amazing
- 40:59job and like Aptify just gives you thousands of scrapers.
- 41:03It's kind of a cheat and sell to your question because it's kind
- 41:05of like, you know, wishing for more wishes because it's just
- 41:07like thousands of scrapers. But that's that's just such an
- 41:09awesome one. I really like the YouTube one,
- 41:12actually. OK, Yeah.
- 41:15And what about like voice agents is would that be 1?
- 41:17Oh yeah. I don't really think of it as
- 41:18like an integration. I think of it as a as a
- 41:20capability of the best one, but. OK, OK.
- 41:23What part of all this is still too hard for Lindy, but that you
- 41:26think Lindy will be able to tackle?
- 41:28That's a good question. I think of Lindy's today and AI
- 41:31agents. They're not really AI employees.
- 41:34Like that's more aspirational. I think of them more as like AI
- 41:36interns. So any simple enough tasks that
- 41:40you can define as a sequence of steps, right?
- 41:44Even if it's always the same sequence, you can give a
- 41:45playbook to someone and ask them to do it.
- 41:48Like AI agents are really good for this.
- 41:50It does like a lot of ambiguity or if it's really a problem
- 41:53where it's like, look, I don't really know, you've got to
- 41:55figure it out. Like when you hire VPS at a
- 41:57startup like AVP of sells, AVP of marketing.
- 41:59I love this quote from Steve Jobs.
- 42:01That's like, we don't hire smart people to tell them what to do.
- 42:03We hire them so they tell us what to do, you know, And so if
- 42:06you really need this person that's like, hey, I need you to
- 42:09help me figure out the marketing strategy and like drive it home
- 42:13like this kind of long time horizon task AI agents or today
- 42:18are poorly suited for. But it's changing.
- 42:20It's changing rapidly. Very rapidly, I think.
- 42:23I think in a couple of years it's basically going to be a
- 42:25full blown AI employee. OK, what feature request are
- 42:28people asking for the most? So it's embarrassing we don't
- 42:31have team accounts yet. Like you can't really like
- 42:34collaborate with your team on Indeed.
- 42:35That's that's coming soon. They're asking for computer use,
- 42:38the ability for indeed to use the computer that's coming soon.
- 42:42They're asking for it's a paying for people to create their own
- 42:46Lindies because you've you've got to do that.
- 42:48So just a sneak peek here. This is just my account because
- 42:51I'm a admin. This is not yet available in
- 42:54Lindy. But like we all building this
- 42:55like it's basically like cursor for Lindy.
- 42:58It's like a an agent copilot that creates agents for you.
- 43:02Yeah, man. OK.
- 43:05What did I miss? Is there anything I didn't ask
- 43:06you that I should have asked you?
- 43:07Like my mind is blown. No, I think we can build a lot
- 43:11of good ground. It's it does make your head spin
- 43:14to think of AI agents like today and then we're also going to be
- 43:18in the future. I think people don't totally
- 43:20understand what's happening right now.
- 43:22Well, we know where to find you. Well, we know where to find
- 43:24Lindy. Where can we find you?
- 43:25Just e-mail me at a flow at Lindy dot AI.
- 43:28I'm also on on Twitter at X. Do people say X now?
- 43:32Altimo ALTIMOO. OK.
- 43:36And we know that if we e-mail you, we're probably going to get
- 43:38Lindy responding, but that's OK. You never know.
- 43:40But yes, we'll be responding. No, I do.
- 43:44I do read. I do read the most.
- 43:45Just a million nails. Yeah.
- 43:47Well, thank you for your time. Yeah, thank you so much for
- 43:50having me. I would really appreciate it if
- 43:52you used my Lindy affiliate link in the show notes.
- 43:54I'll get a little kickback. You'll get a little kickback.
- 43:56And please share it with a friend.
- 43:57We'll see you next time on the Kerner office.