YASH POOJARY (Every / Sparkle) - Reverse engineering apps

=== YASH POOJARY (Every / Sparkle) - Reverse engineering apps ===

(06:59:38): now uh yash joining us yash is the gm of sparkle what up hello can you hear me we

(06:59:45): can hear you tell us where you are right now because i know you're in a really

(06:59:48): special place i am in uh in bocos del toro which is uh an island uh i'm an island

(06:59:56): boy for a few days

(06:59:59): I have been following recommendations given by our CEO here and having a lot of fun time.

(07:00:05): I was just writing on ATV 30 minutes ago.

(07:00:08): So I am pumped and excited.

(07:00:11): I was tuning in.

(07:00:13): Why are you in Panama?

(07:00:15): Because we are having an offsite.

(07:00:16): Everybody is having an offsite in the next few days and they're going to jam

(07:00:22): together in person and create more magical experiences.

(07:00:26): Yeah,

(07:00:26): I don't know if that's been mentioned yet,

(07:00:27): but much of the Ever team is flying to Panama on Friday or Saturday to have an

(07:00:35): all-day week in Panama hacking together.

(07:00:39): And I'm sure we'll be doing live streams and launching some stuff from that.

(07:00:45): So stay tuned for that.

(07:00:46): But we're not here to talk about that.

(07:00:48): We're here to talk about Sparkle.

(07:00:52): What do you want to talk about?

(07:00:55): I want to talk about... Introduce Sparkle too.

(07:00:57): Yes.

(07:00:58): Let's do that.

(07:00:59): Let's do that.

(07:01:00): I was just tuning in to Naveen's stream as well.

(07:01:04): He had this beautifully organized desktop.

(07:01:07): Everything is in place.

(07:01:09): Nothing was scattered.

(07:01:10): So I'm going to share my stream.

(07:01:13): And my entire screen.

(07:01:17): Okay.

(07:01:17): So you see this, right?

(07:01:19): This desktop is perfectly organized.

(07:01:22): I didn't do that.

(07:01:23): I built an app to do that.

(07:01:25): So I'm with Sparkle.

(07:01:26): Sparkle is a Mac file organizational app, which we've been building for the past year.

(07:01:32): And we're super pumped that we have a bunch of features that's coming up, for example,

(07:01:37): which is deep clean.

(07:01:38): It's going to help you save up space on your Mac.

(07:01:41): So this is just a demo.

(07:01:43): Our launch is for the first week of March.

(07:01:45): So you get like information into this.

(07:01:48): You can look at duplicates.

(07:01:49): We rebuild a lot of the systems and figure out,

(07:01:51): okay,

(07:01:52): why organization sucks on your Mac and how we can do a beautiful job with it.

(07:01:58): uh like there are a bunch of things like you're figuring out okay the files that

(07:02:01): you know at 10 000 files on my mac which i've just like downloaded but never used

(07:02:05): and we're just gonna make more of this available to you guys very soon uh and

(07:02:13): that's about sparkle yeah why um so we've reached like just a little quick history

(07:02:19): about sparkle sparkle was first built by dan in like 2023 i think like

(07:02:25): maybe when ChatGPT, the API first launched, and it was an Electron app.

(07:02:32): Then we

(07:02:33): then we rebuilt it and I actually designed it and we launched it.

(07:02:39): It was still an Electron app.

(07:02:41): Then we brought you on and you rebuilt it for a third time and you made it not an

(07:02:47): Electron app because that was really bad.

(07:02:49): And then you're rebuilding it again now.

(07:02:52): So I know that like one of the reasons why you're rebuilding it is we've kind of

(07:02:56): discovered two things,

(07:02:59): which I want to share,

(07:02:59): but then I'm curious for you to share.

(07:03:01): One of them is,

(07:03:04): It's without like looking at every single,

(07:03:07): literally looking at the files themselves,

(07:03:09): it's really hard to organize stuff just by file name and metadata.

(07:03:14): But it's really expensive and super slow to like look at every single file in depth.

(07:03:20): um and our computers just aren't there yet to do locally to do that locally so that

(07:03:24): was like that's one problem that we have organizing files like that is super

(07:03:30): organizing files well just based off of metadata and title can be tough but

(07:03:34): probably the bigger thing that i feel like you've uncovered which is why you've

(07:03:38): just built sort of like this declutter feature is um

(07:03:43): The reality is we're like organizing so much crap on people's computers that is such a waste.

(07:03:48): Like I have 7,000 screenshots that I'll never look at again.

(07:03:52): Maybe you can talk about like the latter.

(07:03:54): So why have we decided to clean up?

(07:03:57): Uh, thank you.

(07:03:58): Uh, thank you for that.

(07:04:00): Uh, and yeah, that is true.

(07:04:02): Like you organizing 10,000,

(07:04:04): 20,000 files on your Mac and we just ran into this problem of like,

(07:04:08): it was really slow.

(07:04:10): And a lot of times you would be like, dude, this is screenshot.

(07:04:13): I don't really need it.

(07:04:14): and earlier we took this conscious call of like hey we shouldn't be deleting stuff

(07:04:19): like sparkles should be just moving fires but we then we'll okay let's build this

(07:04:23): thing for ourselves like what do we hate uh you know on the mac and for me it was

(07:04:29): like screenshot when i learned this for the first time i had 7 000 screenshots and

(07:04:32): now just like removes it uh immediately as it pops up but that's how i like it a

(07:04:38): lot of people use it like once a day and just

(07:04:41): uh we went back to like looking at Marie Kondo and how she looks at like

(07:04:46): organization and tying up stuff and it's just like organization just like two

(07:04:50): things and 70 of it is just throwing away things you don't need and especially if

(07:04:56): you've been using a mac for like years uh or decades uh in my case i just have so

(07:05:01): much stuff i just don't need

(07:05:04): A lot of these things are very level one things.

(07:05:07): But if I go into my clutter,

(07:05:09): it's just like 10 gigs,

(07:05:12): which I just can get back within a few clicks.

(07:05:15): And that's a philosophy that throw away stuff you don't need and the stuff that's

(07:05:19): remaining after that,

(07:05:20): let's build a great organization system.

(07:05:23): And we're going to start looking into content as well.

(07:05:27): So, and you can start prompting as well.

(07:05:29): Like this is structure I want.

(07:05:30): This is, Hey, I don't like this.

(07:05:32): So those are the things that we are building towards.

(07:05:35): Yep.

(07:05:35): Cool.

(07:05:37): Um, well, I know you have something that you want to walk through and share with everyone.

(07:05:40): So why don't we jump into that?

(07:05:42): Awesome.

(07:05:43): Uh, install, um, excellent.

(07:05:47): Okay.

(07:05:47): Uh,

(07:05:48): I just showcased you a demo of the new version of sparkle,

(07:05:52): and I'm going to share with you how I built stuff.

(07:05:55): and how I learn from like million dollar companies.

(07:05:58): And a lot of it, it has to do with like reverse engineering, like macOS apps.

(07:06:05): And I outsource my R&D.

(07:06:07): So I'm going to tell you my secret.

(07:06:09): My secret is that I, Steve,

(07:06:14): But I don't steal code.

(07:06:16): I don't do copyright espionage.

(07:06:18): James Bond can do that.

(07:06:19): I have more important things to do,

(07:06:20): like making sure your organization system on your Mac is perfect.

(07:06:24): And how I do it is I'm going to show you.

(07:06:26): But more importantly, why is that needed?

(07:06:30): is if you look at any writer, they've read thousands of books before writing a book.

(07:06:35): Kobe Bryant, Michael Jordan, they study hours of game film just to prepare.

(07:06:42): And before you start your own restaurant, you go stay at another restaurant.

(07:06:46): And same for an artist, you just copy Picasso stroke by stroke.

(07:06:50): It's not to become them.

(07:06:51): It's just to learn what excellence looks like.

(07:06:54): And I started asking the same question, like, hey, I am a developer.

(07:06:57): I want to become better.

(07:06:58): This is my craft.

(07:06:59): And how do I get better at my craft?

(07:07:01): And that's where the whole question of like,

(07:07:03): dude,

(07:07:03): I have apps on my Mac,

(07:07:06): which are like,

(07:07:06): we have thousands of developers and like billions of dollars invested into this.

(07:07:11): So I can sort of like reversing those apps and use that as my guiding compass while

(07:07:17): building new stuff.

(07:07:19): Uh,

(07:07:20): so yeah,

(07:07:21): what I'm going to quickly show you now is like,

(07:07:23): okay,

(07:07:23): what's that secret prompt that I use,

(07:07:25): like reversing any of these apps?

(07:07:27): How do you actually use, like take advantage of these things?

(07:07:31): And what are these like big picture concepts you should be looking at?

(07:07:35): Uh,

(07:07:36): I'm going to drill down the like reverse engineering,

(07:07:38): like chat GPDs,

(07:07:39): uh,

(07:07:39): the Mac OS apps,

(07:07:40): Spotify,

(07:07:41): DNA,

(07:07:41): a couple of other apps,

(07:07:43): if you have the time and also show the process of like how you can start

(07:07:49): incorporating this into your workflow.

(07:07:52): Okay.

(07:07:53): Now, okay.

(07:07:54): If you want to build a chat app for your Mac or, uh, on your Mac.

(07:07:58): Going to Cloud Code and going like, hey, build me a chat app is very December 2025.

(07:08:04): We are already in the third week of January.

(07:08:08): We don't need that.

(07:08:10): So let's get into it.

(07:08:12): So I have this,

(07:08:13): I follow this person called Kevin Chen,

(07:08:15): and he wrote this blog in like 2022,

(07:08:18): which is about reverse engineering the Rewind app.

(07:08:21): Like Rewind was a company that recently got acquired by Facebook.

(07:08:25): He basically went down into this rabbit hole of like looking under the hood.

(07:08:29): And I always found this fascinating.

(07:08:31): of how is an app made?

(07:08:34): I was very curious.

(07:08:35): So he basically listed out his entire method over here.

(07:08:39): And now you can use this method as your prompt to reverse engineer other apps.

(07:08:44): So if you open any macOS app,

(07:08:50): Like if I go to chat GPT, it's actually just like a bunch of files and folders.

(07:08:57): Like when you look at it,

(07:08:59): you might not get the context of it,

(07:09:01): but an LLM is very well wired to figure out what's in it.

(07:09:06): So let's do it.

(07:09:07): Let's reverse engineer this.

(07:09:11): yeah maybe somebody could find that that this uh article too and drop let's drop

(07:09:15): this in the chat for everyone yeah i'm gonna i'm gonna we're gonna provide the link

(07:09:19): for this uh i'm gonna drop this uh

(07:09:22): I can drop it in the chat.

(07:09:24): Don't worry about it.

(07:09:25): I know somebody that's listening to this.

(07:09:27): I'm going to do it.

(07:09:28): OK.

(07:09:29): OK.

(07:09:29): So there are four steps to how can you do it.

(07:09:32): You need the article link.

(07:09:34): We'll get you that.

(07:09:35): You need to give the actual app on your Mac.

(07:09:38): That should be important.

(07:09:39): You need to ask the LLM to save it to an MD file so that you can use that learning.

(07:09:44): And you just need to do that process over and over.

(07:09:47): OK,

(07:09:47): so without further delay,

(07:09:50): OK,

(07:09:51): I have this prompt ready,

(07:09:54): which basically says that reverse engineering the chat GPT app using this method.

(07:09:59): You just add in the blog.

(07:10:01): And you just go to your set applications folder.

(07:10:09): And you can drop in any app that you want.

(07:10:11): I am doing it with ChatGPT.

(07:10:13): And just say, save the learnings to a new .me file.

(07:10:18): That's it.

(07:10:18): That's all it takes.

(07:10:20): And then I'm going to hit Enter.

(07:10:22): It's going to go to the research.

(07:10:24): OK.

(07:10:25): See, it ran an error.

(07:10:27): And I knew that would happen.

(07:10:29): And that's why, like any cooking show, you always have the dish ready.

(07:10:33): So let's look into this.

(07:10:37): So what it does over here, it's going to tear down.

(07:10:40): Before you do that, what was the error, just in case people run into that?

(07:10:44): It's just like a chat.

(07:10:45): This isn't an error.

(07:10:47): Oh, yeah.

(07:10:48): It's probably going to be like, yeah.

(07:10:50): It's an internal classic.

(07:10:52): Demo day error.

(07:10:54): Got it.

(07:10:55): Yeah.

(07:10:57): And I'm just going to run down this document with you.

(07:11:01): yeah okay so what we're looking at right now is like a complete teardown of how

(07:11:06): chat gpt which is a hundred billion dollar product with like the best engineers in

(07:11:11): the world have built their mac app so we're going to learn from now from this now

(07:11:15): how they do it so well

(07:11:17): Yeah.

(07:11:18): Yeah.

(07:11:19): Let's say you want to build a Mac chat app for macOS.

(07:11:22): You can use this as a guiding principles.

(07:11:25): You know, okay, what are the privacy permissions that are required?

(07:11:29): What are the frameworks they're using?

(07:11:31): If you see this live kit, this is actually a different startup.

(07:11:34): So they're not building their own tech.

(07:11:36): They have outsourced that.

(07:11:38): Okay, they're using screen recording.

(07:11:40): Okay.

(07:11:41): Oh my God, look, they're using Sparkle.

(07:11:43): Yeah, but that's a different sparkle.

(07:11:46): Yeah, see, they're catching on to it.

(07:11:50): And you can sort of start looking into what they are using and start asking why are

(07:11:54): they using it and how you can start using that as well.

(07:11:57): Like if $100 billion company like Brandon said is using it, you should be using that too.

(07:12:03): Because a lot of time, what your speciality is going to be is the experience.

(07:12:08): A lot of these things are solved problems.

(07:12:10): You should not be wasting your time on that.

(07:12:12): You see, a lot of this is like, how do you parse Markdown?

(07:12:15): OK, there's a package for that.

(07:12:17): You might not get a lot of this initially, and that's completely fine.

(07:12:21): But you can ask an LLM.

(07:12:22): You sort of also get like, what are the API endpoints?

(07:12:26): It's going to go really deep, and that's the beautiful part of the prompt.

(07:12:30): If you go manually into Cloud and be like,

(07:12:33): hey,

(07:12:33): reverse it into this app,

(07:12:35): it's not going to do this great of a job.

(07:12:38): And you can sort of start looking into it and go deeper and deeper.

(07:12:42): This is an exhaustive document.

(07:12:44): And what I use it personally is for scoping things.

(07:12:48): When you start building something new, you don't know what the scope is.

(07:12:51): You can ask Cloud or Chai GPT, but it doesn't have skin in the game.

(07:12:55): When you look at a billion dollar app, they have tested this out.

(07:12:59): And if they're using it, you just use it.

(07:13:03): And that's sort of my sort of workflow principle.

(07:13:07): And it does not have to be a ChatGPT clone.

(07:13:11): Even if you have like anything which is like image generation app,

(07:13:14): you can sort of look into that because ChatGPT has image generation,

(07:13:17): even though it's not an image generation app.

(07:13:19): And you can start learning how they build a lot of these things.

(07:13:24): And that's it's like reading a book.

(07:13:26): You can have like Walter Isaacson,

(07:13:28): like writing a novel on like Steve Jobs and Elon Musk on how they think.

(07:13:33): Right.

(07:13:33): And you now get like every app on your Mac is a book and you can open it and start reading it.

(07:13:39): I think what's really amazing about this is if you want to build a chat app,

(07:13:44): this is a good one to go do a teardown on.

(07:13:47): But if you want to build a specific feature in your product that there's really no

(07:13:52): comparison to,

(07:13:53): or your product doesn't have a comparison,

(07:13:55): you can still go to another app that has a similar feature,

(07:13:58): do a teardown of that feature to understand how to build it in your product.

(07:14:02): So it doesn't need to be one-to-one relation.

(07:14:05): 100%.

(07:14:07): Just for fun, I tore down the Spotify app.

(07:14:11): This is how I spent my time,

(07:14:14): just looking at macOS apps and just trying to figure out why did they use this.

(07:14:18): And what I found out was Spotify is actually not a native app.

(07:14:23): And that's very interesting to me,

(07:14:25): because they're using a lot of web technology,

(07:14:28): and they've packaged it into a Mac app.

(07:14:31): even at that scale and what sort of native is i think their audio framework which

(07:14:36): should be native so this all figured i think what's great about telling these downs

(07:14:41): is you can sort of start asking okay what's the 80 20 over here what should i focus

(07:14:47): on when i'm building this new beautiful app where should last part of the energy we

(07:14:53): direct it to

(07:14:54): and that's going to save you a lot of time and if you also have uh you know

(07:14:59): questions about like uh how like okay should i do this and not do this you can just

(07:15:04): tell claude that hey just go see how spotify implemented it and then get back to me

(07:15:09): and this sort of becomes your

(07:15:12): learning document.

(07:15:14): You don't have to ask like the internet or you don't have to like,

(07:15:17): like in the old days,

(07:15:18): you would have to go to Stack Overflow,

(07:15:20): get like go through so many opinions.

(07:15:22): But those are opinions.

(07:15:23): They have not seen scale.

(07:15:25): What you need is like a sort of like you need a direction.

(07:15:29): And these documents are like great at that.

(07:15:32): So maybe you're going to get into this,

(07:15:33): but what's an example of how you've done this,

(07:15:35): uh,

(07:15:36): with sparkle?

(07:15:37): Yeah.

(07:15:37): Yeah.

(07:15:38): Uh, okay.

(07:15:39): Uh, yeah.

(07:15:40): I've done this like a lot of times,

(07:15:42): like,

(07:15:42): uh,

(07:15:43): if there's a package,

(07:15:44): uh,

(07:15:45): we have to use for a lot of stuff,

(07:15:46): like packages are basically code that's written by someone else and you can sort of

(07:15:51): integrate it into your app.

(07:15:53): Uh, sort of like look at Legos and like Lego that you can attach.

(07:15:56): And, um, like there's a lot of like calculation.

(07:15:59): Like when I, uh,

(07:16:02): I'm going to show an example.

(07:16:08): When you have to calculate duplicates, it's a very expensive process.

(07:16:13): It's very technical.

(07:16:15): And you have to look at, is this file similar to that file?

(07:16:19): Is there a minute difference?

(07:16:20): And I used a package for this.

(07:16:23): And I looked into another app which had built the algorithm.

(07:16:27): And I could integrate that.

(07:16:29): And it saved three days of dev.

(07:16:32): and in like i could do it in like three hours and i feel that's like the beautiful

(07:16:36): part of it uh like this is how i do any feature right now like i have like a in my

(07:16:42): sparkle repo i have like a bunch of apps which are like already like when the

(07:16:47): teardowns are ready and i just sort of like plug in play and ask law to like look

(07:16:51): into that and then get back to

(07:16:53): yeah all right so you're learning from the greats and you're you know that you're

(07:16:56): spending most of your time figuring out um what feature should i build next and

(07:17:01): what is the experience going to be that yeah really resonates with the user but

(07:17:05): actually building it either cloud code can do that or you have some amazing

(07:17:09): comparison that you're doing a teardown of that will show you hey this is literally

(07:17:13): like the packages or the frameworks you need to use to actually build this 100

(07:17:18): Yeah.

(07:17:18): And you can just sort, like, I always ask, like, what did they build and what did they buy?

(07:17:22): That's sort of a guiding principle.

(07:17:25): And like, what did they prioritize?

(07:17:27): Those things are very important because as a developer,

(07:17:30): like in large engineering companies,

(07:17:33): you have like engineering managers.

(07:17:36): And one of the responsibilities an engineering manager has to figure out you're not

(07:17:39): doing things you should not be doing.

(07:17:40): You're not spilling up your hours.

(07:17:43): And because we have a lot of these tools which are so accessible,

(07:17:46): like within one prompt,

(07:17:48): which is going to run for 15 minutes,

(07:17:49): and it's going to give you back code,

(07:17:51): you're going to get this false feeling of your building.

(07:17:54): But you're not.

(07:17:56): As Brandon said, you should be focused on the experience.

(07:17:59): Like, for example, Sparkle.

(07:18:02): like we build this but we have a lot of work to do for the experience and that's

(07:18:05): where all of our energies are focused on right now and you need to be you need to

(07:18:10): start doing that and i also love the way that you build yash where like you know

(07:18:16): obviously sparkle right now like it it's there's zero design effort that's been put

(07:18:20): into to yeah um and like it's very different from the way i build like i have a

(07:18:24): designing background and i get really stuck on the way something looks

(07:18:28): But you are just dead set on making sure that the functionality is there.

(07:18:33): We're going to figure out all of the design after the fact.

(07:18:35): And I think that's a really good methodology.

(07:18:38): I just look at it like you're making a movie.

(07:18:41): There's a scripting phase.

(07:18:42): There's a research phase.

(07:18:44): Then you're going to shoot.

(07:18:45): And then you're going to edit.

(07:18:46): So I'm always in like,

(07:18:48): OK,

(07:18:48): if you're shooting right now,

(07:18:50): I'm not really concerned about making the corners perfect.

(07:18:54): We're going to do that in the edit phase and then we're going to like market it

(07:18:58): like how we do it at every.

(07:19:00): But if you want to take like one thing,

(07:19:01): just pick one app that you admire and reverse engineer it.

(07:19:06): It's just so good to read.

(07:19:08): And I feel like it's just like such a big hack that just sort of left out.

(07:19:13): We build like software so much, but we just don't learn from each other.

(07:19:18): like uh and that's something i think we do internally at fb a lot like we learn

(07:19:22): from each other's like workflows apps like we all have access to each other's

(07:19:26): depots but you can start doing it like with in out there in the world cool um all

(07:19:33): right so my understanding is cloud code is down right now um yeah do you have a

(07:19:37): demo that you want to show and is that going to mess mess up no

(07:19:41): Uh, no, uh, I don't have a demo.

(07:19:43): Uh, like it's sort of broke.

(07:19:45): So what it will just run for like 15 minutes.

(07:19:48): Uh,

(07:19:49): and it's going to give you that file that I just showed you,

(07:19:52): uh,

(07:19:53): 9% of it is just the prompt and then you're going to,

(07:19:56): okay,

(07:19:56): now it's running.

(07:19:57): See,

(07:19:57): if you see,

(07:19:58): like it's going to fetch it,

(07:20:00): uh,

(07:20:01): and then it's going to just give us what we already had.

(07:20:04): Um, and okay.

(07:20:06): And that gets us to our next topic.

(07:20:08): Uh, let's, let's get focused.

(07:20:11): And there's a scene which,

(07:20:12): like,

(07:20:13): sort of inspired me to become a developer,

(07:20:14): which is,

(07:20:15): like,

(07:20:15): from the social network where Eduardo walks in and,

(07:20:19): like,

(07:20:19): Mark Zuckerberg is,

(07:20:20): like,

(07:20:21): coding on his laptop.

(07:20:22): And he sort of uses the term, like, wired in.

(07:20:24): Like, what does that mean?

(07:20:26): And it just, like, you're in a flow state, right?

(07:20:28): Like, all day, every day.

(07:20:30): Like, that's why coding is so... Coding used to be meditative.

(07:20:33): Like,

(07:20:34): you can just look at an editor and just,

(07:20:35): like,

(07:20:35): forget about all your concerns and just,

(07:20:37): like,

(07:20:37): write code.

(07:20:39): uh i used to love that and there was a transition transitionary phase uh between

(07:20:43): where i was not writing code lm's writing code i'm like oh my god that sucks

(07:20:48): because i used to love that part that's my playtime he just took up in my playtime

(07:20:53): and i remember we were talking about this uh in turn you get every and like dan put

(07:20:59): screenshot and tweeted it and then Reid Hoffman commented on that that what should

(07:21:04): you be doing when agents are running and he replied with run no agents and i'm like

(07:21:10): okay that's that's true right like Reid Hoffman is saying it's probably going to be

(07:21:13): true and that's where i sort of like started asking uh asking this question like

(07:21:18): how to get better and like managing agents because you're not you're reviewing code

(07:21:22): not writing it there's a spell spelling error fix it

(07:21:26): Um, and what started happening is like the LNM would spit out code.

(07:21:31): I would be like, okay, let's go to the next step.

(07:21:33): That's the wrong mentality.

(07:21:35): What if you train yourself is for like,

(07:21:37): you need to be responsible for every line,

(07:21:40): every line that's there in your code,

(07:21:41): you need to read it.

(07:21:42): Uh, that's sort of like a shortcut.

(07:21:44): It's like, I feel like it's like TikTok, TikTokification of code is like, like swiping up.

(07:21:50): That's wrong.

(07:21:51): That's not how you build technology.

(07:21:53): And it's okay.

(07:21:53): We are figuring it out.

(07:21:54): It's very early days.

(07:21:57): So you need to make sure that you're reading it.

(07:21:58): Are you saying you review every line of code?

(07:22:00): Yeah.

(07:22:01): Yeah, I read it.

(07:22:02): I read every line of code.

(07:22:05): I do want to say that this is,

(07:22:08): you know,

(07:22:09): I'm like,

(07:22:10): I'm sure Kieran is listening to this and I know Kieran doesn't read.

(07:22:14): Sure, he reads some of his code, but he doesn't read a lot of his code.

(07:22:17): Why do you do that?

(07:22:22): I do that because I started writing code because I love it.

(07:22:29): It's like reading a book.

(07:22:32): I don't need a reason to that.

(07:22:33): It just makes me feel, OK, I'm understanding everything.

(07:22:37): And when something goes wrong, I can fix it in two minutes.

(07:22:41): I had the same conversation with Naveed, and he reads every line of code.

(07:22:46): We're using Codex and Cloud,

(07:22:48): but it just was reaching a point where I was unfamiliar with my code base.

(07:22:54): I didn't know this thing that I am working on day in and day out.

(07:22:57): And I would just feel like, OK, that's not how I want to build stuff.

(07:23:01): I need to know exactly what's happening.

(07:23:06): And it's sort of like learning to drive.

(07:23:08): It feels weird, especially early on, like leading core, like managing agents.

(07:23:15): But you're going to figure it out.

(07:23:17): It's going to become a part of your workflow.

(07:23:20): And just so that I stay in this flow state, that's a large part of

(07:23:25): what I like to be in while building is I built a scene called Agent Watch,

(07:23:30): which is sort of like it's an app.

(07:23:32): You can download it.

(07:23:32): We provide the link.

(07:23:33): You can just search Agent Watch on Google,

(07:23:37): and you'll get a GitHub link,

(07:23:39): and you can download it.

(07:23:40): And if you see what it does, it shows all the agents that are currently running.

(07:23:44): So green is running.

(07:23:47): Orange, it's done running.

(07:23:49): Green is done running.

(07:23:50): Orange is sort of running.

(07:23:51): So this agent over here, if you see the chat GPT is running.

(07:23:54): it allows me to switch in and come in and back and forth so i have an agent running

(07:24:00): and see if you see it just went to green so just subtle notification in my status

(07:24:05): bar that hey this agent is done running so i have like two or three cloud code

(07:24:09): agents running at an even point of time

(07:24:11): And that allows me to stay in this flow state.

(07:24:14): Like I'm not on YouTube or like Instagram.

(07:24:17): I'm just like in the zone of like reading my code and like managing my agents.

(07:24:23): And this is sort of a, this is like part of my workflow.

(07:24:26): And like, uh, how I look at it is like,

(07:24:29): I want to get from point A to point B. I'm going to write the code.

(07:24:33): I can't compete with that.

(07:24:35): But my job is to give direction.

(07:24:38): And even though Google Maps is going to show me that,

(07:24:40): hey,

(07:24:41): this is five minutes or 14 minutes I walk,

(07:24:43): I still need to make that journey.

(07:24:46): I still need to evaluate, manage, and review.

(07:24:49): i still have to review it like i can't be like okay go do this and like it's gonna

(07:24:54): come back with something which is not up to my standard and for me this is how i

(07:24:59): enjoy what i do in like 2026.

(07:25:01): this was very different in 2022.

(07:25:03): yeah that's it i have spoken cool man i feel exhausted yeah

(07:25:12): and also i just i mean i do want to say you built agent watch at our last think

(07:25:16): week that's correct which makes me wonder what's going to come out of this next day

(07:25:20): next week um anything in your mind that you're thinking about building i have so

(07:25:25): many things uh like something about like ban um like our agent native narrative i

(07:25:32): feel that's very interesting so how is it in sparkle outside sparkle i feel that's

(07:25:37): something i'm figuring i'm just really excited to

(07:25:42): like hang out with everyone in person.

(07:25:44): Yeah.

(07:25:45): I mean,

(07:25:45): I think one thing that we realized about Sparkle too is like everybody wants their

(07:25:49): files organized differently.

(07:25:50): Yeah.

(07:25:51): And like,

(07:25:52): you know,

(07:25:53): obviously we are building the ability for you to chat with Sparkle to create it,

(07:25:58): you know,

(07:25:59): a custom folder structure.

(07:26:00): And I saw some questions in the chat about like,

(07:26:03): what about,

(07:26:04): you know,

(07:26:04): if I have a repo somewhere and I don't want Sparkle to organize that,

(07:26:08): all of that is being handled in this next version.

(07:26:12): and you know you have the ability to sort of chat with the agent and say hey this

(07:26:18): file is really important like i don't want you to ever touch this i want to be

(07:26:21): manually organizing this all that's going to be sort of in this new version but if

(07:26:29): there's one thing that you could like snap your fingers and have built out uh for

(07:26:33): sparkle what would what would that be sort of like a big big opportunity type

(07:26:38): feature

(07:26:39): Uh, in the current version or like just like.

(07:26:42): Generally for sparkle that you're like, I really wanted to do this.

(07:26:46): Yeah.

(07:26:47): I genuinely want to figure out like our consistent cleanup.

(07:26:51): organizational value because i feel like when i start my day i don't want to see a

(07:26:56): cluttered desktop or like a cluttered downloads folder like it's just very personal

(07:27:01): when i see like hundreds of files over there which i've downloaded and it just

(07:27:05): trash a lot of times but sometimes sometimes it's not and we sort of like want to

(07:27:09): figure out like what's important what's really not important because it's all about

(07:27:14): i want to feel productive when i open my laptop and start coding and i just feel

(07:27:17): like this is so important and a bunch of

(07:27:20): like the spark user i've interacted with have this similar feeling and that's like

(07:27:24): we want to hone that like how do you be more productive on your mac like like your

(07:27:29): disc is running out while you're working on a presentation is something you should

(07:27:33): not have to give your attention to like we have agents for god's sake you should

(07:27:37): like do it in the background just let me know and i think that's something like how

(07:27:41): you build this productivity suite on your mac with this one app and so you can just

(07:27:47): do what you are good at

(07:27:49): Yeah.

(07:27:49): What do you think about search?

(07:27:50): Do you think search is something we're going to invest in or something that we're

(07:27:53): going to deprecate?

(07:27:55): I think it's just like,

(07:27:57): if you're not going to be number one or number two at something,

(07:28:00): why should we do it?

(07:28:01): I think that's why, like, if it's not a hell yeah, it's a no.

(07:28:06): And I feel it might be like, okay, we might do it separately.

(07:28:09): Like, we might be the best search tool on your Mac.

(07:28:12): And right now,

(07:28:12): it just feels like what we are headed towards with Sparkle,

(07:28:16): which is like files and stuff like that.

(07:28:18): That's such a big...

(07:28:20): market or like that's such so much work to be done and so it's sort of like is

(07:28:25): interesting but it's sort of an accident like i don't see it fitting in perfectly

(07:28:30): in our new vision but yeah we might do something which is like best search uh best

(07:28:35): search file search on your mac

(07:28:38): Yeah, I think something that we talk about a lot at Every is what are you going to sacrifice?

(07:28:43): Because we're all so excited about building different things here.

(07:28:47): And we all come up with ideas every day.

(07:28:50): And you cannot build something great without like,

(07:28:53): focusing on it for a very very long time yeah so um you know regularly we are

(07:28:58): asking ourselves what do you sacrifice if you're going to do this what are you

(07:29:01): going to sacrifice to do it and it's worth it sacrificing that yeah um yash you're

(07:29:07): the man can't wait to hang out with you uh in a couple days uh and uh thank you so

(07:29:14): much for taking the time thank you for having me this was so fun

(07:29:17): Awesome, man.

(07:29:18): Awesome.

(07:29:18): All right.

(07:29:19): I'm not really sure who's coming on next, but I know.

(07:29:21): Oh, Natalia.

(07:29:22): What's up?

(07:29:23): Hey.

(07:29:24): Brooker.

(07:29:25): How are we doing?

(07:29:26): Doing great.

(07:29:29): I'm going to leave it with you guys.

(07:29:30): Thank you, everybody, for sticking around.

(07:29:33): All right.

(07:29:34): Thanks, Brandon.

(07:29:35): All right.