# Perplexity Computer: Autonomous Agents for Founder Productivity

**Podcast:** The Startup Ideas Podcast
**Published:** 2026-02-27

## Transcript

Today we're going to take Perplexity Computer for a spin.
Now, I haven't tried it, so together we're gonna basically look at how can we use this?
How can entrepreneurs use this to make money and be more productive?
We're actually gonna ask it to tell us what are the top use cases to use, and we're gonna test it out and see if it's any good.
You've probably seen Perplexity Computer, and you might have bookmarked it, but today we're actually gonna see if it's super valuable.
So let's just get uh right into it.
The startup bag is podcast.
I'm gonna just say, hey, I want to know the five best use cases, non-obvious for uh founders who want to make money and be more productive using perplexity computer.
Like, what are the five prompts and use cases I need to be aware of that are super mind-blowing.
I'm actually going to share this live on YouTube and Spotify on the Startup Ideas Podcast.
So we are going to test it.
This is on uh perplexity computer.
Uh, as of now, you need to be on the max plan to to get it going.
Uh, that's $200 a month.
They say it's coming to pro and maybe flee free pens.
I'm not sure.
I'm not affiliated with perplexity by any stretch, and this is the first time I'm using this.
So uh just noticing uh for the first time, you basically have these tasks on the left hand side.
Um, and on the right hand side is what's happening uh with the different agents.
So it says, let me load up the relevant skills to give you the best possible answer here.
Let me read the key reference files to make sure I'm giving you the most accurate picture of what a computer can do.
And it's running all these tasks in parallel, which is pretty cool.
You don't need to open up a bunch of different cloud code instances.
It's right there running it in parallel.
So now it says, good.
I now have a deep understanding of the full capability set since you're sharing this on the Startup Ideas podcast.
Let me craft five generally non-obvious founder specific use cases that will map make your audience jaws drop.
All right, let's see if perplexity computer can drop the sauce.
Here are five non-obvious use cases that surprise should surprise your audience.
These are built for founders who want to move faster and make more money.
Warm outbound at scale.
So personalized prospect reach research and outreach in uh in one shot.
So let's go ahead and try this.
We're actually you can, I guess you can press copy right over here.
And I guess you want to start a task.
So here's a list of 30 companies I want to sell to.
For each one, find the founder, CEO on LinkedIn, research their company's recent news, funding, and pain points, and then draft a hyper personalized cold email referencing something specific about them.
Send each email via my connected uh Gmail.
So we're gonna go ahead and do that.
So that's actually pretty cool.
You have your tasks over here.
You can go back and forth.
Um, I think that's pretty clean.
Perplexity's always done a good job at doing pretty clean uh UIUX.
Uh sip of cold brew.
Let's get keep it going.
All right, I'll tackle this step by step.
I need a few things to make this effective.
Again, load it running the task in parallel.
I see a couple things I need to address before we get execute on this.
Gmail is not connected yet.
I need to get this uh set up so I can send emails on your behalf.
I need some key details from you to make these emails actually convert.
So this is cool.
I mean, connecting your email, uh, you know, there's definitely obviously some risk associated with that, but I'm gonna go ahead and do that.
Um, so I've gone ahead and connected it, so it's super cool that you could just do that pretty easily.
Gmail is connected, I'm ready to roll as soon as you share the list of 30 companies, paste them in your reply, what you're selling, your SaaS product, podcast sponsorships, tone and CTA preferences.
Once I have those details, I'll research all 30 companies in parallel.
I'll find the right person at each and draft hyper personalized emails and send them through your Gmail.
So, you know, I'm gonna say uh I run the startup ideas podcast.
This is for podcast sponsorships.
I want you to research who advertises on TBPN, and then I and I'll say take the top, let's just say five companies, and then you now know I'm selling pod sponsorships and tone and CTA.
Well, tone, I want to be friend to friend, clear, calm, and CTA.
Do you want to set up a call with my team?
Let's go ahead and do that and see uh, you know, if it's able to find who advertises on TBPN.
So it says uh I need to find the first company to advertise on TVPN.
So it actually doesn't know what uh does it know what TVPN is?
Um it does.
So it's first says the Bitcoin podcast network.
It's not called that, I think it's called Technology Brothers Podcast Network, but it still finds the right one over here.
You can see uh because you can see that it's uh it's five it's finding a bunch of uh like TVPN signs with CA, that's the right TBPN.
So it founds it finds them.
So it had a little trip up over there, um, and then it just says um it finds their five yeah, ramp, plaid, Figma, and it's researching all five in parallel, recent news, pain points, and any specific I can reference an email.
So this is really interesting, right?
Because uh, I don't know about you, but I actually don't do any cold email really, unless it's to a cold DM of someone I I look up to and I think is really interesting and cool.
I'll I'll just cold DM them.
But for like closing leads, finding leads, stuff like that.
I don't do cold DMs.
My approach is really building audiences and communities and having people reach out to me.
Um so this is cool that it you know I can do this.
Um, but what I've had issues with in terms of cold DMs in the past is it just sound it sounds lame, it doesn't sound like me.
It's a lot of work to actually find the people.
Um you know, let's just see if this is this is any good.
So it says, excellent research.
Let me find the right contact emails and draft the emails.
Let me also look up who handles sponsorship and partnerships at these companies.
Since emailing the CEO directly about a podcast sponsorship isn't always the best path.
This is what I'm talking about.
This is this is really cool.
Um because like obviously Dylan Field, the founder of Figma, isn't gonna open a cold DM, probably from from my email.
Um but like the podcast sponsorship people might, you know.
Um, so it says good.
I have the send underscore email tool ready.
So what's really cool is uh perplexity computer, you're spinning up a computer, it's like a virtual computer, it has skills, it has tools, um, and it can actually go and do things.
So let's see if this is any good.
Um, okay, so I here are the five personalized emails, each referencing specific details for my research.
So it's actually crazy.
So it found Toby, the founder of Shopify's email.
Um that's really cool.
Like that in itself is really, really crazy, right?
Um, so good on them for finding that.
But we did talk about how we shouldn't uh uh let me check if these CEOs are reachable.
Oh, interesting.
So it's actually I wouldn't send it to CEOs personally, so I guess that's something that you know we can adjust.
But actually, I mean, I would probably send it to the CEOs and I would send it to the podcast people.
So let's just read the Toby email.
Hi, Toby.
Been watching Shopify's run lately, 11.6 billion in revenue.
I think it's more than that.
I think maybe not actually.
30% growth and the AI catalog tools are generally impressive.
The MRI thing you did with Cloud was wild.
So I don't know if you saw you saw that.
He he he did like um I think he analyzed like an MIR result and built a tool around it, and it was wild.
So it went and checked his X account and what he's been posting.
I run the Startup Ideas podcast, our audience is founders and builders actively launching companies, many of them choosing their commerce stack right now.
We get about X downloads per episode, and our listeners skew heavily towards early stage founders who are making infrastructure decisions.
I noticed Shopify is already sponsoring TVPN, which makes a lot of sense, or shows it's a similar demo, but a tighter focus on people who are literally in the process of starting companies feels like a natural pit.
This is this is awesome.
And then there's another one to ramp, um, and there's another one to plaid, and they have all these people's email addresses.
I don't know how they're getting them, but um, assuming those are real email addresses, that's pretty pretty cool.
I'm gonna say uh so, and by the way, and if I press send email, uh, it would just send the email it's got connected to my email to my email address and go ahead and do that.
But what I'm gonna do, I'm gonna do one more thing, and then we're gonna get to the next use case.
Um just curious if it can do this.
So I'm gonna say uh CEO's emails are impressive, but realistically, these are busy people.
Let's try to get in front of the person in charge of podcast creator partnerships.
Doesn't that make more sense?
And then I'm gonna just give it a bonus question.
I'm gonna say, by the way, is there something I'm not asking you here that could make me more money in this use case?
I'm just gonna see what it says.
Sometimes, you know, we're trying to push it to the limit.
This is where we're reviewing perplexity computers, so let's uh let's go and see.
You're absolutely right, CEOs aren't reviewing sponsorship pitches.
The person who actually signs podcast deals, usually head of brand marketing partnerships, and it goes and says, Let's go.
What uh, you know, let's go find those people.
So it's gonna go ahead and find them.
It also says what you're not asking, but should be outreach to companies that sponsor TV VN's competitors too.
Right.
So if a company sponsors TVPN, they also sponsor my my first million, all in acquired, right?
Set up a recurring monitor.
I can check TBPN all in and other tech podcasts weekly for new sponsors.
The moment a new brand starts advertising on competitors' brand, you get a notification with their partnership contacts contacts.
So you're reaching out while their podcast budget is hot.
That is so smart and so cool that you can set it up.
That's like, you know, one of the reasons I think people like open claw and and co-work now has some scheduled tasks.
Is you can yeah, schedule a task.
So it's very cool that you can spin up a perplexity computer and have these recurring monitors and go and do that and do things for you in the cloud.
Uh follow-up sequences, one email converts at maybe 10 5 to 10%.
Maybe I can, or I can schedule follow-ups three and seven days later with new angles if they don't reply.
Absolutely crazy.
Want me to do any of those after we get these five out the door?
Duh.
Um, okay, so let's just see.
Um good findings.
Now I have solid leads.
Let me compile what uh what I found.
So it's gone ahead and found them.
It's found.
Wow, it's found their email addresses.
It's hey George, caught your 20 VC.
This is the VP of growth of RAM.
The caught your 20 VC episode.
The bit about finding alpha and channels, no one else is trying to really resonate.
No one else is trying really resonated.
Direct mail becoming one of Rant's biggest channels, such a good example of that.
Like these emails are amazing and they're so personalized.
We're just gonna I I am impressed here.
Uh I'm going to I'm gonna say I'm just gonna copy this.
What you're asking, and not should be yes, go do those things.
And after that, I promise we will get to the new use case.
But I will say this is this is uh better than I expected, a lot cleaner.
Um I like the UI.
I like that you can spin out multiple tasks um at the same time.
Um it is 200 a month.
So is that expensive or is that cheap?
You know, depend if you're I guess if you're using it like this, like obviously this would pay for itself in this in this actual use case.
Uh it's going and researching sponsors uh across competing prods.
It's setting up weekly monitor that checks those pods for new sponsors and sends you notification.
What I like about the Lindy AI assistant, and I'm gonna do a video, I think, uh just completely on that, is you can text it out using iMessage.
So uh I don't know if you can do that here.
I don't think so.
Um but I do think that like one of the one of the cool things about open claw has been that you can use iMessage, you can use Telegram, you can use uh Discord.
Um so you know uh this is a nice UI, but I do like the idea that like I can just text it um on some of those other platforms.
Okay, so oh wow.
So the email has been sent, which is scary because I kind of know some of these people, but we'll see what happens.
I'll I'll report back and tell you know, if you see ramp, plaid, Figma, Shopify App Lovein advertising on this podcast.
Uh it hopefully came from that uh personalized email.
So I'll keep you posted on that.
Um what's happening now is it's updating a plan with expanded outreach, it's identifying top five TBN sponsor companies and their CEOs and founders, it's researching all five companies in pat in parallel, finding the partnership, brand marketing contact at each.
It's researching sponsors across tech podcasts to build the bigger prospect list, and it's doing follow-up sequences and weekly monitoring in parallel.
This is literally a marketing email marketing person's old job, point blank.
This is and it's using uh Cloud Sonnet 4.6, which is great.
I will say I wish it didn't send those emails without me clicking send.
When I went up here, uh, if we scroll up here, it was like, do you want to send these emails?
And I didn't click send.
Like, I actually didn't think it was gonna click the emails.
I'm all for it because I think it's fun to see like if you know if it converts, but I you know would have loved to to have to have double checked that.
So this is really cool.
It's researching all the sponsors.
Um let's go ahead and while the it's it's sort of running here.
Let's go ahead and do some more use cases.
Um automated competitive intel, the second second one here.
Automative competitive intel on autopilot, recurring monitoring with push alerts.
Every morning at 8 a.m.
Check these five competitors' websites for pricing changes, new feature announcements, and any new blog posts.
Also search X for any mentions of them.
If anything has changed since yesterday, send me a push notification with a summary.
And if nothing has changed, stay quiet.
Why is this mind blowing?
This turns a computer into persistive competitive intelligence agent that runs every day without you touching it.
It browses their actual websites, compares it with a found yesterday.
Okay, let's go and see if this works.
We're gonna head and start a new task.
I'm gonna go ahead and you can see what's cool.
You can also just dictate here, but we don't need to.
And I could see myself getting used to this.
I don't know about you, but I can I can definitely see myself getting used to this.
Which competitor should I monitor uh monitor?
Okay, so you know, let's just say we're let's go with the the podcast stuff.
So let's just say my first million.
Um, what other podcasts?
All in podcasts.
These are the top three, and you figure out the other two.
My pod is called the Startup Ideas Podcast, and I'm Greg Eisenberg.
Okay, let's yeah, let's get let's get a perplexity computer to do some of the work.
Why do I have to do all that work?
So uh I'm just testing it a little bit.
Um, says remembering Greg's podcast and identity, so it's updating memory.
So that's really cool.
Um, just loving, I'm loving this.
I'm loving this.
Uh just watching it cook.
It it feels really good.
Based on my research, too strong competitors are 20 BC and acquired, both target the same entrepreneurial audience and frequently overlap with your listener base.
Cool.
Now let's set up the daily 8 a.m.
monitoring tasks.
I need to convert 8 a.m.
Eastern to UTC first.
So it's going ahead and like coding, which is really cool.
Um, so it's going ahead and coding.
Um, and that's why, you know, it's a computer with access to you know, 16.
I think it's 16 different models.
We saw before was using Sonnet 4.6.
Um, and it's supposed to be using the best model for the best task.
And uh, you know, it's just interesting.
It's interesting watching it, how how it thinks.
So every morning I'll automatically check these five competitors.
What I'm watching for pricing response for the changes on their sites, new feature announcements or product launches, new blog posts or content, new episode drops, guest announcements, notable ex activity and mentions.
If anything has changed, you'll get a push notification with the summary and links.
If nothing has changed, you won't hear from me.
First run is tomorrow morning at 8 a.m.
Uh, I picked a 20 VC and acquired as the two additional.
Yeah, we know that.
Okay, so I'm gonna say that's cool, but what would today's report look like?
I want to see it so I can give feedback.
So because you're not gonna be there with me at 8 a.m.
I just wanted I want you to see what you know what it's capable of.
So perplexity computer goes, good call.
Let me run the full check right now so you can see exactly what tomorrow tomorrow's report would look like.
So if this works and you're a founder of or you want to be a founder, you want to build something, there is no reason why you shouldn't be having something like this going, right?
Um I wonder if you can I wonder, like here.
So let's scroll up a bit.
My question is so you're gonna run this, but can you email it to me, right?
I'm gonna ask that.
Like, I would love for it to be emailed.
I don't necessarily want to like go and check perplexity every morning at 8 a.m.
But I much rather it just be sent to my email or communicated to me.
And you know, like I said, I what I liked about Lindy AI is you can use some and and open clause, you can use iMessage.
So it's using Sonnet 4.6 again.
It's looking at all the latest uh episodes of the podcast, doing research, compiling the reports, and then uh, oh, that's cool.
You can go and click and see all the subtasks.
This reminds me a lot like man of Manis.
Um really good job at that.
And I actually think Manis is just like underrated as a platform.
I really need to go back in there and do another episode on Manis.
Um yeah, they it's going ahead and doing that.
It's cooking, it's cooking, it's cooking.
Um yeah, I think I think this uh I mean, I I would open this, you know.
I would open this and and not because I want to do what like quote unquote competition is doing on any business I'm starting, but it does help like get your creative juices flowing, and you know I'm all about getting uh creative juices flowing.
I find it helpful even if you want to be the anti-version of them, right?
Sometimes a competitor will launch a new feature or do something, launch a brand campaign, and it's not about hey, I want to do what they want to do.
It's like actually, like I, you know, I want to be different than them.
Great example of that is uh, you know, open AI announced they're doing ads, anthropic announced they're not doing ads.
They're the anti-version of each other.
And I think that uh I think having a competitor analysis report, having agents go ahead and and doing this for you, uh, sending it to you could be really cool just to help you get those creative juices flowing and get the most out of your day that way, right?
Like, if you can, I think that's like a big part of what you know, why a lot of us want to use AI agents is like if we can get unfair advantage in terms of information, help us frame that information, help to get the most out of that information, then uh, you know, we'll be we'll be ahead.
So let's see what's happening here.
So it's actually saved um it's saved it as an MD file into uh home slash user MFM report.
Well, I guess the report for each of the different uh podcasts, the all in TBPN, um the 20 VC, and so it it creates those files and then it's done.
And this is what so this is what the notification would look like.
A new episode drops today.
Uh so it says new episode.
This is Harry Stebbings.
Anthropic wipes billions off the market.
Um, it's the new episode.
Published a European microfunds report, confirmed as a speaker.
I mean, I'm just scrolling through this, and like it's interesting.
My first million, Ryan Smith, you know, from the Utah Jazz is on there.
That's really cool.
Um, Pat Walls, dumb iPhone apps making money.
That's cool.
Andrew Wilkinson, best worst business in 2060, very cool.
Sam Parr pivoting YouTube channel away from personal finance towards founder-led content.
His second video already hit 100,000 views.
Like that in itself is interesting, right?
No format or pricing changes to MFM.
It's like these are this is interesting.
Um, and then it says that's a format on quiet days when nothing is meaningful or change across all five.
You get no notification at all.
What do you think?
Anything you'd want added, remove or formatted differently.
How are you going to send me this report?
So I like these reports, but like, how am I gonna get them?
Right now it's a push notification through perplexity.
Interesting.
So you have to download the app.
Um so what is the best way?
Yeah, I guess you could.
Um, could you send via email?
Is that even possible?
Can you send via iMessage?
I don't think it's gonna be able to send via iMessage, but you know, let's.
A man can dream.
You know, we can dream, we all can dream.
Let me check what's messaging and email tools are provided are available.
So it's gonna go ahead and do that.
Email's ready to go, emails already connected.
I can send the full report to your inbox every morning.
iMessage isn't available.
So, anyways.
Uh it gives the, you know, it it could send via email, which is really cool.
Um, okay, don't send emails for now.
All right.
So let's go ahead and see what our cold email stuff is up to.
Wow, this is this is crazy, man.
Oh my god.
So it's researching sp uh sponsors across competing pods.
It's setting up a weekly monitor, checks those pods for new sponsors, send you a notification with partnership contacts, and it's scheduling the follow-up sequences at day three and day seven.
Who doesn't reply?
It's just doing it all.
It's found 96 sponsor product prospects found.
This is crazy.
I bet you this can, I mean, this this is this is something that could definitely work.
All right, let's go back and do another use case.
Uh investor pipeline research, deep dive 50 VCs before you fundraise.
Here are 50 fit fifth 50 VC firms.
I'm considering raising four for my series A for each one.
Find their most recent fund size, partner, blah blah blah.
Put it all into a spreadsheet.
Why it's interesting, computer runs batch research across 50 firms in parallel, web, search, social and x, linkedin.
Uh you get back a structured spreadsheet with everything.
I'm kind of I'm not that interested in raising money from VCs, but I I am interested in seeing how this could work and how you can create spreadsheets.
So we're just gonna test to see how this is able to do some research and throw things into spreadsheets.
Um it says I need to clarify a few things before diving into this research.
So it says I need some details to make this actually target and useful.
Which company you're raising series A for.
Let's pretend.
Let's pretend I'm the CEO of well, I'll just say I'm the co-founder of idea and browser.com.
The number one website for people to find great startup ideas, trends, and how to build them with AI.
I'm at series A level.
So I'm just giving it the name of the company.
Um, and it's gonna go ahead and hopefully create that spreadsheet.
While it's doing that, let's go ahead and see another use case.
Turn a podcast episode into a full content machine.
Whoa, that's cool.
So here's my latest podcast recording.
Transcribe it with speaker labels, uh, write a 15-word box, 100-word blog posts, pull out the five most tweetable quotes.
That's really interesting.
So it's cool that you can do that.
I'm not gonna do that because I don't have uh an audio file handy.
But basically, what you can do is you can set up so that every, you know, every time, like you can set up a recurring workflow, I guess.
Like every time you launch uh a new podcast, uh, you can extend it to, you know, blog posts, tweetable quotes, LinkedIn carousel, and just like based on how this is working, it just feels like it would do a good job.
Um, okay, so the status here, if you see to the left, it shows that it needs something from me.
Um now it's just asked me, can you share the VC serms?
I don't know who would fund this.
You do the research on the best ones.
So we're just gonna make it do a lot of the work.
Um and let's see, fair enough.
Let me find the best fit VCs for idea browser series A.
I'll load the relevant skills and build a targeted list based on your profile, AI powered platform.
And yeah.
So it's going ahead and says idea browser sits at the intersection of several hot sectors, AI powered tools, creator uh economy, community platforms, and I'm researching across those vectors.
So that's exactly how like a human being would go and uh actually reach out to to venture people.
Of course, if you know them, um, you would just be like, I'm gonna reach out to Jimmy because Jimmy is works at this venture fund.
But if you don't, you the way you would do it is you would say, like, okay, what are VC firms that have are investing in this, you know, genre type of company?
Um, so it's going ahead and doing it for us.
It's using Sonnet 4.6.
Uh, I just expanded this.
It says it's you know, looking at Bessemer and Union Square Ventures.
Um, it's it's researching, it's just going through tons and tons of different articles.
Um, saving the final curated 50 VC firms to workspace.
Um, let's see here.
If you click files, what happens?
Okay, nothing here really.
You know, these are all the connectors, I guess.
So this is where you can use uh, yeah.
We've already connected Gmail and Calendar, but you can connect, you know, Google, Google Drive, OneDrive, Dropbox, Slack, um, Hrefs, you know, uh, Cloudflare.
I'm just looking at the ones that are interesting to me.
Google Gemini, HubSpot is really interesting.
Um, yeah, because imagine like your CRM, right?
Your CRM has all the data, whatever you use, even if you use like Google Sheets as your CRM.
If you have all the data and you can load it up into here and you have a computer that can go and access this, that's really cool.
PayPal, uh, sending money, that's crazy.
Reddit, that's cool.
Um, you know, how do you find pain points and stuff like that?
You gotta get you know, ideabrowser.com on here.
All right, waiting for your response.
So it's gone ahead and identify them.
Um interesting.
This will search far and wide across the internet to get you fund sizes, sector, recent investment, best fit part.
This will consume a significant amount of credits.
Do you want to proceed?
At least it asks.
At least it asks.
Um, I'm not sure how many credits you get with um your max plan.
So uh, you know, I've been using this for now like 30 minutes.
Um, so good, but we'll see, right?
I will say that even like if this costs 200, like this session, it wouldn't would have been worth it for me.
Um, because again, like you close a sponsorship and it works out, that's worth it.
Okay, so it's gone ahead and researching all in parallel.
And it this is wow, this is this is crazy.
These are all like, yeah, these are people I would reach out to if I was raising money.
So this is this is really cool.
All right, let's while it's doing this, let's go ahead and check our last uh use case live market diligence on a deal.
A full investment memo from a single tip a ticker.
Build me an investment research memo on Shop.
Hold their latest, that's Shopify, pull their latest financials, earnings, transcript highlights, compare their margins and growth to big commerce and Wix.
Check what analysts and finance tweeters are saying compile into polished PDF, but chart, I want bullcase, bear case, and your assessment.
I've been seeing, I will say, I've been seeing on X a lot of people using Perplexity Computer with the use case of using it as like a financial analyst.
Um I saw uh I think I think it was I think it was Morgan Linton who built like a Bloomberg terminal uh use like vibe code using perplexity computer, uh, which is really cool.
So it seems to be like really primed for financial um financial uh advice.
Wow, this is actually this is cool.
So it says I'll build a comprehensive investment research on Shopify.
Let me load the data visualization and research assistance skills.
So it's cool.
You don't have to like install the skills, right?
It's going ahead and doing that for you.
Um, it knows, so I do use perplexity, and it knows that uh I like like value investors, and Bill Ackman is a value investor, so it's gone ahead and like brought that in.
So it looks at your old searches on perplexity, and even though I've never used perplexity Computer, it knows a little bit about me, which is interesting.
For people who use uh perplexity, that context will already be loaded.
So it says I have all the data I need.
Let me get the shop estimates and analyst research, then build the PDF.
I have plenty of data.
That's a little but you it's a little flex there, Perplexy Computer.
Let me now update the tasks and build a comprehensive PDF with charts.
Let me download the price history in CSV and build a complete PDF with charts using a sub agent.
So what's cool about you know what I'm learning?
What's cool about computer is subagents, skills, tools, that's all built into this.
And people thought perplexity was dead.
Here we go.
The CSV CSV downloaded uh as single lines.
Let me save the data properly from the content I already have.
All right.
So you can see the status here.
It's working.
While it's working, let's just ask it.
Um you have any more use cases people should try.
I'm live right now.
Just curious if it has any of them.
Uh, but this is something that you should totally do.
Just you know, ask, ask it for use cases.
Uh five more different angles.
We won't have time to do today, but just you know, could be really interesting for you.
Steal your competitors' SEO strategy, auto-generate a branded pitch deck from a loom video.
That's pretty cool.
Shadow CFO, weekly financial sanity check on autopilot.
Amazing.
Hiring sourcer, find and rank 50 candidates a minute, the deal room, due diligence package on an acquisition target.
More.
Give me give me a few more.
Those are all really interesting.
I think the hardest part of a lot of this is like figuring out okay, which agents recurring flows should I have.
Reverse engineer any SaaS pricing page.
Wow, that's cool.
You instantly see exactly where you underpriced.
That's really cool.
Customer discovery on autopilot, free customer research.
Clone your best performing ad copy.
Wow, this is crazy.
So yeah, let's check in on this.
So okay, so here's here's what we got.
It's got the name, the fund of this is the VC, the VC list, the partner's name, recent tweets, recent interviews, the different sources.
That's crazy.
And it's building a yeah, it's just building an Excel spreadsheet with all that data.
Literally building an Excel spreadsheet.
It's an analyst.
Alright.
Safe to say this out this exceeded my expectations.
I don't know about you, but this is this is really cool.
Something I'm gonna be playing more with.
Um really interested in sort of the open clawification of the internet.
So this whole this whole world where now you can spin up these machines uh with agents, sub agents, tools and skills, have it access files, and get them to do things for you.
I think a lot of us want to start these one person, $1 billion companies, want to smart small teams or even a small team and building these small companies, software companies, agent companies, SaaS companies.
And that's why I like playing with these tools.
And you know, I'm just I'm grateful that you know you come and and and you and you you watch these videos, you listen, you like, you comment, and you know, I'm sharing everything I'm learning in real time.
Um and I just find this just such an exciting moment in in entrepreneurial history.
Um, I built and sold, you know, three startups, advisor to TikTok and Reddit.
I've invested in a ton.
Um, been doing this since I've been a kid, and um yeah, just I just find like right now just to be so fun to tinker, have fun, and build assets.
So I'm excited to see what you think of this uh perplexity computer, and I'll be reviewing and and checking out more of these uh in the future because I think uh it's an unfair advantage.
So um hope this has been helpful.
Um send it to a friend if it if it has been, and I'll see you next time on the Startup Ideas podcast.
Um have a creative day, my friend.