🚀 Add to Chrome – It’s Free - YouTube Summarizer
Category: AI and Business Automation
Tags: AI AgentsBusiness AutomationCustomer SupportSales AutomationWorkflow Optimization
Entities: AI agentscomputer useFloGregLindy AILinkedInMacintosh of AI agentsZapier
00:00
I just saw the future of AI agents, and I'm not even overselling it. I had the founder of Lindy AI show us new stuff around how to use plain English to do vibe marketing to to build out AI agents
00:16
that make you money and make you more productive. A lot of people have promised this.
A lot of people have promised AI agent platforms that do this, but it is finally here. And this opens the door for fully autonomous
00:31
companies that are run by agents. And that was science fiction.
Um, you know, literally a few weeks ago, it's not anymore. I can't wait for you to see how the new Lindy AI works and how it uses
00:47
agent swarms, something called computer use, and it's going to change the game. So, tune in.
[Music] When it comes to AI agents, Flo, you're
01:05
probably one of my first calls that I make. And today's an extremely special episode because Flo, what are we going to learn today?
We are going to talk about the two major things we're announcing today which are uh the agent builder and computer use.
01:20
Um so the agent builder makes it super easy to build the agents and computer use makes the agents basically able to do pretty much anything that you can do on a computer. So I I I think like we've been talking you and I Greg about AI agents for for a while now like and and they've been sort of immature like these
01:36
two things like like a big big big step forward on with AI agents. Okay.
Because there's a lot of episodes you know on AI agents on YouTube and X including some of mine where I'm like or or me and people are like this is insane this is insane Flo is this actually
01:51
insane but if people stick to the end of this episode will they be like my mind is blown. Look, I don't mean to sound too grandiloquent like the tech is new, right?
That that's the reality of it. I think we're like, you know, what we are driving as an industry towards what we're talking what what we're calling
02:07
the the Macintosh of AI agents, right? So, we've been stuck in like the terminal days of like personal computers is like, yes, it's insane.
Yes, it's really good, but like it's really hard to use. And I think like that's one of the main concerns people have about AI agents.
It's just super hard to use and
02:23
to create and leverage this new technology. So yes, this is a really really really big step forward because now you're going to be able to create really complex AI agents really simply.
It's like you just talk to it like, "Hey, I want you to do X moving forward and and this and that." And you just
02:38
have a conversation and it's effectively an AI agent creating an AI agent for you. The two concerns people have about AI agents are number one, too hard.
Can't figure it out. I'm not an engineer.
I don't have the time for that. Number two, it doesn't integrate with my tools.
Well, basically solving
02:56
90% of both problems today. Uh, so look, you know, we don't have AGI yet, but yes, it's it's a really really big step forward.
Okay, you have my attention. I'm intrigued.
All right. Um, let me show my screen then.
So, I'll talk about marketing, I
03:14
can talk about sales, and I can talk about customer support. Those are like three really big use cases.
Uh, support is is straightforward. this like just yesterday I was catching up with a friend uh and he's spending like $12,000 a month on customer support and and and this is after optimization.
He's got a
03:30
bunch of like VAS in the Philippines and he's in the process of replacing all of that by Lindy. And until now he was holding back because he was like, "Yeah, but they do so much that's like it's it's more than just answering a ticket." Like, "Yes, we've been doing the rag thing like the retrieval thing.
Like,
03:46
yes, you know, if you ask a question about a refund policy or like a return policy, we can answer that. No problem." But very often they check your order status on my Shopify or like they issue a refund on Stripe or like you know there's all sorts of things that that that it does.
Um so now with computer
04:02
use like you can't do it just like okay you receive a support ticket or they asking for a refund. You insert a human into the loop probably at that point you're going to receive notifications every like bus I'm about to issue this refund.
Yeah. And then say yay and just goes on Stripe like hey check the order
04:17
status on Shopify goes on Shopify can do all of that stuff for you. So support is just like such an easy one, especially for like e-commerce businesses.
Like if you have a little bit of seasonality, it's not just about the the the money saving. It's also like it's just easier to operate a business that's backed by
04:33
AI agents than a business that's backed by humans. Because yes, AI agents are not perfect, but at least they're consistent.
Like if you've gotten it to work once, it is going to work a thousand times. Well, as humans, you never know.
And they don't scare like they quit on you. They start slacking
04:48
off. they start like sometimes stealing from you.
It's, you know, like AI agents, you can just put it on autopilot and forget about it and then move on to another part of your business to focus on. So support is just like a total no-brainer.
Sales, ah, we have so many
05:03
sales agents. So, I'll start with like a huge caveat like I want to be real like and I I think that that's one thing I really love with your podcast Greg is like I feel like there's so much again like grandiloquent promises out there that are like empty promises and I feel like the AISDR frankly is one of them.
05:20
Like if you think you can just spin up an AISDR agent and click a button and then infinite pipeline is going to come through your door like you're out of your mind. Like this is just not how it works.
like it is still a lot of work if you don't have an adbound motion like even if you hire the human SDR and you
05:36
don't have an adbound motion you should you should expect like I think 40% chance that they never succeed and even if they do succeed you should expect a couple of months of like heavy iteration and this is not because the human is dub it's just it's hard there's like a lot of iteration but once you have figured
05:52
it out yes like uh AISDRs can be very effective my recommendation is that I think an AISDR is basically supplementing a human SDR. So that was the whole adbound thing.
I think the the way of the future is you should hire a
06:08
person full-time who is basically 10x more effective than they would be if they were alone. But you still need a person full time to manage your AISDR agents.
And so here with with computer use the things that we are unlocking with our AISDRs, man, I feel like I feel
06:23
like I'm going to get banned from LinkedIn, but LinkedIn LinkedIn DMs. So you've got to have a lot of touches.
So, you know, we send an email to people, we send them a text message, we can do that. Uh, if they have opted in, you've got to collect optin.
That's like a legal thing. But if they have opted in,
06:39
we make them a phone calls like the Lindy agents can do that. And then we hit them on LinkedIn.
So, it's just like all of these touches and it all happens inside a single workflow. And LinkedIn actually turns out it's it's the most effective at least for now.
So, outreach is a really big one.
06:57
Nurturing is another really big one. So, you know, we jump on these calls and like a good win rate if you're like if you've got a sales team is like 25 30%.
So, it means like at best you're going to lose only 70% of your deals, but those are actually really good leads.
07:13
Like they jumped on a call with you, they have like the beginning of a relationship and so forth. So what we do is when a deal is lost, we log it in a spreadsheet and then we have an AI agent that nurtures the last deals.
And so
07:28
what it does is in that spreadsheet we log uh the use case that this person had and then this AI agent uh observes two categories of events. It observes did we just win a deal in a similar use case or like in a similar industry.
It's like if
07:44
we're in touch with like a I don't know like a B2B SAS finance company of 50 people and then we close another company that's also B2B SAS finance 50 people like the agent is going to observe that win a we won one we're going to go back to that guy so he looks at the spreadsheets like ah there's one guy
08:00
here I'm going to go back to him like Bob how have you been I was just thinking about you we just closed our business it's like most likely they know about them because they might even be competitors right so now there's like the FOMO right so that's one and the other kind of event it observes is the releases. So again like we we basically
08:16
talk to the agent or like it observes our change logs and so forth and it's like hey okay we're releasing agent builders and we're releasing uh computer use and so we also have that like lost reason in the thing and so it's like ah we couldn't integrate with like we've got a lot of customers in healthcare
08:33
that's one reason I'm so excited about computer use because these guys they have these like a healthcare record management systems and these these systems do not offer APIs like API and stuff and if they do it cost like $100,000 you have to go through like a year-long review process is a nightmare with computer use basically going
08:49
through the back door and so right here like this nurturing agent is going to like looks like aha it's like Bob you know last time we spoke you couldn't you wanted to integrate to your epic EMR computer use just came out like we can do it now do you want to do you want to book a call could we can we build from scratch like
09:07
a full-on agent showing computer use like just from scratch yeah let Let me think of a use case that we could go after. Let's uh let's just do a simple use case that's going to like DM people on LinkedIn.
09:23
Yeah. Uh all right.
I'm going to go here and I'm going to be like when I before you even get, you know, do this. No, 99.9% of people are not DMing people on LinkedIn.
99.9% like, you know,
09:41
including myself. But I know that I if I probably, you know, reach out to people on LinkedIn DM, like I would probably close more business.
I'm just not doing it because like in my mind it's cringe and I don't want to like think about it. But if I could, you know, spin this up
09:58
and and it works, like that's a huge unlock for me. So I just want to just preface it with that.
Um 100%. And I think that's one thing we hear very often is like people are always like, "Ah, yeah, agents are coming for our jobs." Sure, maybe over the long term, but in practice right
10:14
now, what we're seeing is a lot of people who just do stuff they wouldn't do before. Like we know we should post on LinkedIn.
We don't do it because it's really hard to automate and it's kind of cringe and we don't want to think about it. Okay.
So, for now, I'm just going to make it manual. But I could like I could
10:30
be like, "Hey, every time I ping you on Slack or every time someone submits a form or every time someone gets added to this spreadsheet, like hit them hit them up on LinkedIn." For now, I'm just going to be like, um, when I send you a message with a LinkedIn profile,
10:47
uh, I want you to I want you to send them a DM on LinkedIn, um, asking if they would be interested in AI agent services.
11:03
Um, tailor your DM to the person's profile. Uh, and I'm going to be like, I want you to use computer use to send a DM on LinkedIn asking if they would be interested in AI agent sources.
And the
11:18
reason I'm doing that is because we also have LinkedIn integrations that this is the ironic part by the way, like people think, and I thought that API integrations would just be the bomb. They're just like universally better.
Actually, computer use very often works better. So, it's like, hey, okay, so when you send me a message, it's
11:34
confirming. It's like, yeah, you want me to do that and start a computer session and do all of that stuff?
like, yep. So, it's just this is crazy, dude.
Like, you've been building this for a while, so you probably are like numb to it, but the fact that you can just literally talk to it and it's and it's building it
11:49
live like this is the dream. Oh, yeah.
I agree. Uh, I mean, we have been Thank you, Greg.
You're right. I've become like a little nub to it.
But, yeah, I mean, this has been what we've been working towards for the last three years. Like, straight up.
Like this has always
12:05
been the vision is like you tell it what you want to do and it just and it just does it for you. You shouldn't have to be an engineer.
It's just doing it. Dude, you're going to put all the NN bros out of business.
Look, we we do have some competitors
12:21
that I think are unnecessarily complex. Um Um All right, it's done.
I've created your LinkedIn DM at retention. When you send a message with the LinkedIn profile URL, A and if I inspect the agent here is like all right you're going to send
12:36
me a message the message if the message contains a LinkedIn profile URL I'm going to start a computer session and then you are helping to send a personalized LinkedIn DM to a prospect about AI agent services um we are going to redact here because what I need to do is I need to enter and this is not going
12:51
to be here in production but I have a computer ID here um that I need to use for it to have my login you can see it even like named the agent it's like LinkedIn DM outreach agent I'm going to save do this and let's do it. Uh, Greg,
13:06
I'm going to I'm going to spam you, man. Uh, oh, I I liked I liked it when I was DMing, not getting the DM.
Wait a minute. You know the tweet.
It's like realized that Leopard's eating people's faces with eat my faces. Um,
13:22
exactly. Okay, so the LinkedIn the condition passed.
It's starting a computer. We should probably do something in the meanwhile because again the big problem is like it's not a big problem but it's just know this is going to be a pain in the butt.
Yeah. I mean while it's doing that we
13:37
can I mean brainstorm other ideas to get just give people a sense of like what what sort of things that they could be creating. Yeah.
Um oh actually it's not that slow. Let me let me just open the browser and then we can like tab away once it starts
13:52
being boring. All right we did it.
Uh, open phone is my shared phone number with my assistant. I still have a human assistant.
Uh, all right. And so you can
14:08
take control. And when you've taken control, you can then release control and let the agent know like, hey, I just did something.
So, uh, here I'm going to be like, I just logged in. I feel like I'm watching the future, dude.
Like, this is this is crazy. Yes, it's it is.
And again, it's early.
14:25
I always want to be careful to like be real with people, but like it's working. Yeah, it's interesting because I was so excited about chat GBT operator when it came out because to me that was a glimpse of the future, but what it was missing was like the agent builder, the
14:42
workflow, like you know, that's right. There was only so far it could go.
Yeah, 100%. I think being able to plug it to your applications also via APIs and being able to give it different triggers is really powerful and then being able to
14:59
have this like state transition. So in the case of like my ex to LinkedIn like it's actually it's it's it's the same agent but it moves from stage to stage and that makes it a lot more reliable because otherwise like operator is just it's surprising like it's not good enough to do a lot of this stuff even
15:15
simple tasks like this but here you're like holding its hand ever so slightly like you can see I'm not telling it click by click what to do but I'm still like all right step one you're going to check my Twitter step two you're going to check on LinkedIn and if any of these steps is screwing up I can really pinpoint where it's doing that like step
15:31
of this workflow and I can update the workflow. And from a prompt perspective, like do you have any advice on how to create optimized prompts for for Lindy?
No. No.
I just like the agent builder takes care of all of that for you. I mean, you've seen me create this agent.
15:47
It's like it's not rocket science. It's like I just I just talk to it like I talk to an intern and just figures it out.
And again, that's the beauty of it is like, oho, okay, it's found your profile and now it's going to
16:02
Uh, ah, You don't have DMs. You're like in a photo.
Let's see. Let's see if we can figure this out.
But this is a good example of what I mean by like that's the advantage is like you have this double mode of like this double loop of like you can edit the
16:18
agent and the agent instructions and you can see the agent operate and you sort of ping pong between these two modes because if the agent screws up when it operates you can just go in these instructions you're like okay like keep this in mind moving forward. Okay so now it's um it's looking at your like profile to like know how to pitch you.
16:35
Yeah, I disabled messages. Okay.
From random people. I think that's going to become smarter and smarter.
Yeah, it's looking for the message button. I think it's just going to give up.
But it's interesting to see like I want to see what happens. And then we can
16:51
always bug someone else after this. Ah, it's not dumb.
Message Greg is right here. Oh, you know why?
Because I have LinkedIn recruiter. It's kind of messed up how LinkedIn does that.
like you've disabled DMs, but with LinkedIn Recruiter, it's fine because I'm paying for it. So, it's probably
17:07
going to ask me to log into like a LinkedIn recruiter account. Oh.
Oh my god. All right, Greg, are you interested in AI services?
I might be after this DM, man.
17:23
And so, again, I mean, you you saw me create this agent like it took me literally like I mean, we can look at the recording, but like 2 minutes, if that, like two to three minutes or something like that. um it's not going to be perfect first shot like it's working so that's always something but like you're going to want to iterate on like the the the copy that it writes the
17:40
DMs for and and so forth like so I think well and this is what you said earlier like you you know I introduced the human in the loop not just because it's imperfect but also because there's a dose of subjectivity and so I think well that's where we are with AI agents now it's like we're sort of graduating from the phase of like hey it screws up and
17:56
we're moving towards the phase of like now it's just like subjective and sometimes it does lack common sense Wait, dude. This is crazy.
Hi, Greg. Just listened to your startup ideas podcast.
Love how you're always spotting opportunities to quote unquote get the juices flowing in the entrepreneurial
18:11
space. As someone managing a portfolio of internet companies at lay checkout, I imagine you're consistently looking for ways to scale operations efficiently across your ventures.
Matter of fact, I am. I've been working with portfolio CEOs who are using AI agents to automate
18:28
repetitive tasks like lead qualification, customer support, and content creation, freeing up their teams to focus on the high impact work that actually moves the needles. Given your background scaling companies from islands to advising Tik Tok, you probably see the potential for AI to
18:43
handle operational heavy lifting while founders focus on strategy growth. It's true.
Would you be interested in a quick chat about how other portfolio companies are leveraging AI agents? I'd love that.
I'd love to share some specific use cases
18:59
that might be relevant for your portfolio. That's actually really smart, dude.
Are you kidding me? Phil shot.
Are you kidding me? This is this is this is when it's all said, Craig.
Like, you know, like being a founder sucks.
19:16
Yeah. This is the highlight of like it's like you build Kula's product and like and you and you're grinding on it for three years and at some point it clicks and like look it's well like 1% of the way there.
Ah it's in your inbox. Dude, did you did you hear that?
That was someone just messaged me on
19:32
LinkedIn. I wonder who it is.
Um, and so again, now I can like chain it with like I could find your phone number and I could be like, "Hey, send him a text message to be like, Greg, I just hit you up on LinkedIn and if you
19:47
opted in, I could also make you a phone call, right?" Um, and now it could I'm in. There you go.
Um, so I can I could now iterate on this agent to go like, "Hey, if they say they're in, look at my Google calendar, offer them times,
20:03
right? um send them an invite and so forth.
So if you wanted to do that, you would go back to flow editor and just give like prompted. Is that what you would do?
Basically, yeah, I would I would be like, okay, after that you're going to wait for like an hour or something like
20:20
that and then you're going to wait for 12 hours and then 48 hours and something like that. And every time you wake up after waiting, you check if there was a reply.
If there was a reply and if they said they're in, you check my availabilities on my calendar. and you if if and you send and you send
20:36
the availabilities and you go back and forth until you found an availability that works. So this is why like you start really small and then you iterate and that's why I'm saying like look it's it's not like in five minutes one click you're going to have like infinite pipeline but
20:51
you know you sit down for like half a day or so and it's not complicated anymore. It's just a matter of like iterating and in the end I can show you in the end you have I'll show you this example of an agent or this example of an agent.
What else do I have? This example is
21:07
really good. Um as you iterate and again you you build your way towards it but your agents end up becoming really complex and really really really powerful.
So you start small. It's really simple to get started and like this is an agent that they use for recording because like that's how I
21:22
spend half my time and that's what it does. And this is actually not even that complex of an agent.
But little by little to you just keep iterating and iterating and iterating and and you land here. Uh this is one of my favorite agents like my chief of staff agent.
I just like live in this thing all day. Like this is this is one of my more
21:37
complex ones. Um this is my CRM manager agent.
So this one I just talk to all day and you know it logs stuff in my CRM. It retrieves stuff from my CRM.
So that that's where you end up after
21:53
after you've reiterated on your agents. Incredible dude.
You've done it. You've done it again.
I I don't think I've ever had someone on actually no I've never had anyone on the podcast where I'm like this is the future. I need to invest somehow and I need to get involved, you
22:10
know. Um, so that that's I I think the reason I say that is I've been playing with a lot of these tools and they've been close.
Um, but this is the closest thing I have seen to what I had in my
22:27
mind, which is prompt it to to workflows and agent flows and then having uh, you know, a computer go and actually go and fulfill that task. Um so for me what's going through my mind flow
22:42
just to be frank is like you know how do I take stock of you know we have a bunch of companies that we're running like all the tasks that we're doing it how we can productize it and where the where is there an unfair advantage and where there's our opportunity to actually you
22:58
know be more productive and make more money. Yeah I you know I always use the the factorial image.
I don't know if people your listeners are familiar with Factorio, but it's like this factory simulation video game, but I I really do think and I've been using this image for
23:13
years like I really do think the business of the future looks like Factorio. So you end up thinking of your business as like and open a Figma like a fig jam like just whiteboard it out like okay my business is basically this pipeline.
Every business is basically a pipeline and then and then you just take
23:28
stock of like where am I spending most money? Where am I spending most time?
What is the current bottleneck of the business? And let me just overwhelm the bottleneck with AI agents.
I have a top ofunnel problem. I don't struggle to like close people once they're in the funnel, but they struggle to get top of
23:44
funnel. Okay, marketing agents.
I'm going to spin up a bunch of agents that like create a ton of content. We we have a V3 integration that's coming.
So, you could totally have an agent that's like, "Hey, I want you to post a 100 videos on Tik Tok and Instagram every day, right? And then I'm going to give you feedback about the videos." And then you
24:01
you should learn you can very much prompt your agent to be like hey also every day you could have another agent do that like I want you to look back at your videos views history and see what clicks and then I want you to maintain a Google doc with like your learnings so far that you just keep iterating on and
24:16
then the other agent that creates the videos consults this Google doc every day. So you could totally do that.
It's like, okay, I've solved my top my top of funnel problem. Now I have a bottom of funnel problem.
Like people are a pain in the ass to support. I'm gonna I'm just going to overwhelm the bottleneck with AI agents and I'm going to solve
24:32
that and I'm never going to touch a support ticket again. And just keep doing that.
And I look I people don't understand like and I have been careful during this entire conversation to like not be overly granular like so I hope people recognize that like I have been saying for years like the fully
24:48
autonomous company is coming and I don't mean that as like a grandoquent like this is coming in 20 years. I'm like, "No, no, no, no, no.
This this is coming like now, like in the next two years, perhaps. I I I think in the next as early as like 12 months, I do think we're going to see people create fully
25:03
autonomous agentic companies." I mean, when I see when I see something like this, it makes sense, right? Like, um, so of course there's going to be kinks in in the product as you're rolling it
25:20
out. Like, this is the first day that it's live.
Um, but you know, I think that if you have an idea and you have distribution and you have a plan to get more distribution, like you create these like vibe marketing agents and LinkedIn DMs
25:36
and you like what is what is a business if you think about it? It's customer acquisition and fulfillment of the product or service, right?
So, if you of course it's going to be autonomous at some point. Um because if you're if you
25:53
could figure out how to create AI agents that that are, you know, automating the distribution like we've just seen right now and you can figure out how you can create AI agents and automations to do the fulfillment, then why wouldn't there
26:11
be an autonomous company? Yeah.
The the building blocks are here. Like this is no longer a like sci-fi thing.
It's like a Yeah. like this is it's it's it's here.
So like my Lindy recruiter for example like okay this is a Lindy recruiter that I use and so I
26:28
can show you I'm like hey find me I feel bad for the guys at ZPR I like the the guys at Zapier but I'm like hey find me find me 20 software engineers working at Zapier in the US and like all right I'm going to find the people and it's like perfect I found I found 20 software engineers here's the list and if I look
26:44
I can I can look at their LinkedIn and moving forward I'm going to be able to I created this one before we had computer use I should update it to and I DM' them on LinkedIn, but like, okay, this is a bunch of like software engineers at at Zapier. Okay, found them.
And if you look at the timestamps, okay, this was at like 6:30 p.m., okay? And so it's
27:01
like, all right, I found these guys. Who do you want me to reach out to?
Which of these candidates would you like me to reach out? I'm like, yeah, all of them.
This is a crowd list. This is awesome.
Do it. And it's like, okay, I'm going to do it.
It keeps track of its campaigns because like sometimes I get banned on on Gmail because I just sent too many emails. And so it's like, okay.
And then
27:17
we have this feature called like agent swarms. So it's like it just pins up a swarm of like 20 sub agents.
And if you look here, so this was at 6:34. Okay.
So like I kicked off the search at 6:31. 3 minutes later I had emails going out.
Okay, 20 of them. And the first thing
27:34
this sub agent is doing is like it's searching if I've reached out to this guy before. So it's like, no, you've not reached out.
And then it sends an email. And then it wakes up, sends follow-up emails, sends follow-up emails.
It just doesn't let go. So, I have I have the fulfillment part of the equation going.
27:50
I could totally just also spin up another Lindy that like goes on LinkedIn jobs or whatever and looks for companies that are recruiting and sends them a message to be like, "Yo, I saw you were recruiting. I'm an AI agent." Maybe I actually think that would be part of it's like, "I'm an AI agent.
I can help you recruit for like a tenth of the
28:05
price. Just just tell me yes." And you don't even have to sign up.
Like, here's your Stripe link. You don't even have to look at the LinkedIn at the Lindy interface.
Here's your Stripe link. let me know when you've done it and then I'll start like reaching out to people for you.
Dude, thanks for coming on and and and sharing this with us and and giving us
28:22
like the exclusive on this. I appreciate that.
Um I appreciate you being real honestly. Uh there's there's a just not overpromising.
Um and uh how do people get started and you know and start
28:39
building using Lindy? Yeah, you go to lindy.aigreg uh that entitles you to uh like a couple thousand free credits and yeah, you can just get started immediately.
Okay, so if they go to that link, they get a few thousand free credits. So I'll include
28:54
that in the show notes then so people can go there, check it out. Um and uh I'm going to be sharing this with my team, man.
And and I hope to to to build some some cool stuff. Yeah, love it.
Yeah, let me know what you think. And by the way, people like I
29:12
love having Flo on the show. So, um people in the comment section like let me know if you want to have Flo come back on the show to go deeper on some other workflows uh that we could be building together.
So, let me know in the comment section. Would really appreciate that.
Uh and
29:28
Flo, I love Prashan. Thanks, Greg.