July 31, 2025 • 33 mins
Kaya Thomas (kayathomas.is) comes back after half a decade to tell us about how motherhood inspired her new app Milk Diary (milkdiary.com). She talks about using new APIs like Foundation Models, SpeechAnalyzer, and AlarmKit to handle the complex stuff other feeding apps miss: intelligent scheduling and reminders, combo feeding that's actually easy to track, hassle-free tracking for twins and smart milk management.
Guest
Related Links
Related Episodes
- Notifications with Kaya Thomas
- Live from CommunityKit WWDC 2025 with Matt Massicotte
- Plinky with Joe Fabisevich
- ChatGPTovski with Kris Slazinski
- The Making of Callsheet with Casey Liss
- Apples, Glasses, and HAL, Oh My!
Social Media
LinkedIn - @leogdion
GitHub - @brightdigit
GitHub - @leogdion
TikTok - @brightdigit
Mastodon - @leogdion@c.im
Youtube - @brightdigit
BlueSky - @leogdion.bsky.social
Twitter Leo - @leogdion
Twitter BrightDigit - @brightdigit
Credits
Music from https://filmmusic.io
"Blippy Trance" by Kevin MacLeod (https://incompetech.com)
License: CC BY (http://creativecommons.org/licenses/by/4.0/)
- (00:00) - What is Milk Diary
- (04:20) - Foundation Models
- (10:56) - The Feeding App Market
- (13:59) - Liquid Glass
- (17:01) - AlarmKit
- (19:34) - Local and Server Side Storage
- (22:28) - SpeechAnalyzer
- (25:13) - Developing with AI
- Edward Sanchez
- Steven Lipton
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:03):
Welcome to another episode, empower Apps.
I'm your host, Leo Dion.
Today I am joined by Kaya ThomasThank you for coming on the show.
Of course, Leo, I'm glad to be here.
Thanks for having me again.
So what have you been upto since, five years ago.
Pre COVID.
Yeah.
What have you been up to?
(00:24):
Oh man.
it's been a lot in the last five years.
I, took a break from iOS developmentand went outside my comfort zone and
went to grad school and, decided todo some more customer facing work and.
Work on the people side of things.
And, it's been a whirlwind and agreat experience, I think to step
outside of engineering a bit.
(00:46):
And now I'm, trying to mixboth worlds and make a new app.
That's awesome.
So I actually looked it up.
Our episode was like rightat the beginning of COVID,
just to give you an idea.
It was March the 2020, which is hilarious.
The start of it all.
right.
That's when we thought KobeBryant passing away was gonna
be the worst thing of that year.
(01:06):
So Yeah.
It got a lot worse.
So.
Let's talk about thisbrand new app, milk Diary.
You wanna talk about why did you makethis and what did you find, what do you,
what did you find was the need that youfeel like Milk Diary is gonna help with?
Yeah, so I became a mom last year andyou know, that's a huge life change.
(01:31):
And of course, because I love apps.
I wanted to use apps.
As a part of my parenting experienceto help manage and track things.
And so one of the most intensiveparts of parenting, a newborn and
a baby is feeding them and the,
Mental, logistical things you have tokeep track of, and the physicality of it.
(01:56):
But there's a lot to think about,like how often they need to feed.
How much milk do you have for them?
access to them to access, like howmuch you have in your fridge, how
much do you have in your freezer?
Just understanding the differencebetween you feeding them and
you actually making the milk.
So there's a lot of like logisticaland administrative stuff.
(02:17):
And obviously, you know, as developerswe know that the, the great thing about
technology is can help with some of likethe logistical and administrative stuff.
But I was getting frustrated 'cause a lotof the apps, we're not flexible enough and
we're not, really thinking about this kindof full experience and all the different
(02:38):
connections between how you have to makethe milk track, the milk, store, the milk.
Like it's, it's just likethere's this whole holistic.
Journey and process there.
And I felt like all the apps thatI were using might do one thing,
but then when you need somethingelse, you can't do it in that app.
So you might have to use anotherapp or track it another way.
(02:59):
I started to get the itch again andI was like, okay, I think I need to
create an app because, I felt likethere needs to be something better.
And I think that, especially as a mom,I realized that a lot of the apps out
there are not actually made by folks whoare going through this experience, right.
And living through it.
And so having that.
(03:19):
Experience and being able to also callupon, moms, my friends and just moms in
the community and get their experience aswell and be able to really make something
that's catered to the experience ofnew parents and understand what they
would need and how to make an app thatis, user friendly, easy to use, and
actually like saving time and energy.
(03:42):
Well, what does yourdaughter think of the app?
I mean, honestly, She is, I think enamoredwith, technology in general just because
the idea of just touching the screen andhaving something happen, that reaction.
So I don't think she really understandslike, what an app is per se, but
she loves pressing things and,
Ima, yes, I
(04:02):
and seeing it react on the screen.
Yes.
Yeah.
so I asked Mrs. Dion beforerecording, what did she do?
And she just used the notes app.
So, which I'm not surprised.
Or a spreadsheet 'causeshe's loves spreadsheets.
this looks really awesome.
what.
One of the things that you talked abouthere was that really piqued my interest
(04:25):
was the use of foundation models.
Yeah.
what were you thinking as far ashow that would fit in with your app?
Yeah.
So before, dub I was reallystruggling with the idea of like
smart intelligence features, right?
Because I figured thatintegrating some type of.
(04:48):
Intelligence would be great for the appbecause it could help with that more
like efficiency and kind of logisticalthing that I was talking about.
But I was really hesitant to integrateany of the cloud models, like any
of the LLMs that are existing.
Open, open AI and anthropic, et cetera.
Because this data's really sensitive,like I don't think a lot of parents want
(05:12):
this type of data, being sent right overthe internet or being sent somewhere.
And so I was reallyhesitant to integrate that.
And so when they announced foundationmodels, I was like, yes, this
is my opportunity to be able to,incorporate intelligent features
in a really private way on device.
So it was just a huge win.
(05:34):
I think the ways in which I'mintegrating it is a couple ways.
So one is being able to logsessions via text or voice, right?
So all of the apps where you track thingsaround, you have to manually enter right?
Oh, here's how much, and you haveto go to the different text inputs
and, select the different things.
(05:56):
And a lot of times, as anew parent, you're tired.
you don't wanna be like, doing a
have an extra hand 'causeyou're carrying someone.
Exactly.
So being able to just either,you know, input text, or even
speak right into the app and say,okay, like I pumped five ounces.
And then just have it save.
So just being able to save that time.
So that's where the foundation modelsreally come in to be able to have
(06:18):
those prompts that allow them to createthe data, that they want to track.
And I think the difference I foundwith the foundation models really
helped me from a developer side.
I was really overwhelmed.
When I started to think about thecloud models and like how I would
integrate them in a way that wouldreduce hallucinations, get actual data
(06:39):
that, was focused right and accurate.
So the fact that the foundation modelshave guided generation is just so good.
I mean, I think that.
You have to work on tweaking your datamodels to make sure that they work.
So you still reduce hallucinations'cause that's just a thing with LLMs.
(07:00):
But it's much better.
And I think as a developer, it allowsme to have a bit more reliability
and consistency on what I'm gettingfrom the prompts, the user prompts.
So we're using,
I mute Gen General.
General
General, is it a macro or a protocol?
Okay, so you're using that and thenthat would be the output that you're
(07:24):
looking for when somebody sayssomething and that way it's actionable.
Correct.
Yeah, so how I'm doing it and ittook a while for me to figure out
the best way to think about thedata schema for the general structs.
At first I was thinking I would just havea session struct and, you know, it would
have all the data, whether it was a pumpsession, nursing, a bottle feed, right?
(07:48):
And the foundation model shouldbe able to detect, based off what
they're saying, what type of.
Feed it is and be ableto fill in the data.
But I realized actually, at leastin my experience with working with
the foundation model so far, is it'sactually better to have smaller,
structs and have them be very specific.
So breaking them out, becauseI realized that when you have
(08:11):
just like a kind of catchall.
data model for the foundations, thehallucinations increase and it's a bit
harder for it to figure out, exactlywhich variables it should fill in
based off of the context of the prompt.
So you're using like enums and stuffto make it as specific as possible?
Yeah.
Rather than just like a stringwhere you can get something crazy.
(08:33):
so there's both the feedingand the pumping, right?
Are there things you'retracking in the app
Yeah, so it's bottle feeding,nursing, and pumping.
Okay.
And then you also mentioned formula.
how does that fit in?
what's the use case for that?
Yep.
A lot of parents combo feed, right?
So they use breast milk and formulain the app you'll have the ability
(08:55):
to track formula or breast milk.
if you're doing bottle feeds withformula, you'll be able to track
what type of formula you use.
I know a lot of parents also try differenttypes of formula, you can even keep track
of notes to see if there was reactions or
Got it.
things like that.
youngest daughter, so I don't knowif you know my personal story,
but our four oldest are, adopted.
(09:15):
at least the youngest, I think theyoungest girl who's now gonna be 11, she
moved in with us when she was a newbornand we had a lot of formula I remember
doing EXCO development and having her onmy lap while I'm feeding her at 2:00 AM
Oh wow.
I totally remember that.
But.
Formula.
doesn't last as long, right?
(09:37):
As breast milk, as far as I'm concerned.
Yeah.
So, yeah, I could totallysee how that would fit in.
And I also, like one thing you mentionedhere, is, twins or multiples, right?
like having to track multiple children.
That's a really important niche
Yes.
And I have, a close friend who hastwins and I realized in testing out.
(09:57):
A lot of the apps is all of themmake you switch between profiles
in order to track anything.
So you have to create, a profile foreach kid and then if you wanna track
anything, you have to switch, likebetween profiles, to track a feed or.
Yes.
when I thought about like, okay,how am I gonna model this, app?
(10:18):
I was like, okay, from the start,I wanna make it super easy if you
have multiples and you don't have toswitch profiles or do anything, extra.
If you want to just track something andyou just select which kid it's for, right?
So this is totally iOS, 26, correct.
it
iPhone only.
iPhone only.
iPhone only to start.
(10:38):
I think that, if I find that there'sinterest in iPad support or Mac
support, I would think about it.
But yeah, iPhone only
watch support.
Oh yeah, watch support is definitelysomething I want to have because
that is something I used a lotactually when I was feeding.
Not vision os I don't see
yeah,
really useful.
I wanna ask this question, butdid you do like a comparative
(10:59):
analysis before you got started?
Just to see what was currently out there?
And what was themarketing angle for those?
How do they monetize?
Yeah.
Are they basically crappy React apps?
what's the environment out there?
Yeah, it's interesting.
I think so I definitely did a comparativeanalysis and there are, like, to be
(11:19):
frank, a lot of apps out there that,track baby, newborn tracking and,
feed tracking and stuff like that.
But I think, there's definitely,you have the cross platform, popular
cross platform apps that are, in myopinion, I, despise using, because,
I like native apps and I find thatthey were a poor user experience.
(11:41):
I think what I found is.
You have a lot of apps that aretrying to do everything, so they're
like, oh, you can track everything.
Right, but
the thing I was thinking, sorry.
It was just like, oh, Gerberhas a app and it's you know,
it's their app for everything.
And then by the way, we have this littlemilk diary thing over here in the corner.
That's kind of what I was thinkingwas gonna end up happening.
(12:03):
Yeah, exactly.
And then it is the user experienceis super clunky and like, it's
hard to use and know how you eventrack things and where you do it.
I just found it's very overwhelming.
A lot of these apps because they'retrying to do too many things.
And so for me, that's one of thedecisions I made is I really wanted to
focus on the full feeding experienceand be able to support everything
(12:24):
in that space and not try to be thiscatchall that's doing everything.
Because I think that oftentimeswhat makes apps great is when
there is a specific focus, right?
And you're getting really, thevalue out of it because it.
Can help you solve problemsfor a specific need.
a lot of the apps out there, there'ssome free apps out there that have a
(12:44):
bunch of ads, they monetize via ads.
then there's some that are trying tolean into, I think a big area that
some of the larger apps are trying tolean into is sleep tracking and sleep
predictions Personally I don't think that.
Having experienced the ups and downsof a newborn and infant sleep, I
don't think an app can really predict,
(13:05):
Or even like, what'sthe point in tracking?
It's not like you can be like, all right,sweetie, but you can't do the getting up
at 2:00 AM This is what my watch said.
exactly.
I don't, there's nothing actionable.
Exactly.
So like, I mean, there's a bunchthat are leaning into that.
And then like you said, a lot ofthe formula companies have apps
and stuff that I think are, a meansto sell their formula and stuff.
(13:25):
And so, there are a couple of, indieapps that are nice, but I think because.
They are not, like I said, they'revery specific and they don't support.
So they might support one thing reallywell, but they're not thinking about the
full journey of the feeding experience.
And so it's not flexible enough, ifsomething changed about your journey.
(13:46):
So, that's where I feel like I can fillthat space and that gap, to be able to
provide something that's comprehensiveand flexible enough to use throughout
that entire first year from newborn to.
early toddler days,
So when you first developed theapp, were you developing it for iOS?
18.
I was,
Okay.
So.
(14:09):
I wanna really take advantageof everything new, including
foundation models, but also yougot liquid glass to deal with.
How has that gone?
It,
oh
li liquid glass, it's a moving targetbecause every beta, right, like there's
something a little bit different.
I think the most difficult part for mehas been understanding like the best way
(14:35):
to embrace it and where to embrace it.
Do you even have a choice?
for something like tab bars, you don'thave a choice, but when you think about
like buttons and toolbar items and things,how you integrate it, you have a choice in
terms of how it looks and how far you go.
Especially with custom UI elements,How much you integrate, liquid glass.
(14:56):
So I think that's been the hard part offiguring out it's not necessary to have
it for every button And every interaction.
where do you prioritize having it and whatis the prioritization for liquid glass?
Like is it just for primary actions?
It's, so, I think that's whatI've been trying to figure out.
I think the thing that I really don'tlike, about it is, it seems as if
(15:18):
when I look at like the Apple appson, in the beta, right, like notes and
reminders and mail, et cetera, it seemsthat there's a lot of hiding actions
behind menus and, and things thatwhere before at least my understanding
of like the design guidelines.
What's recommended is you wanna makeit easier for folks to know where
(15:41):
to go and what actions to take.
Right.
And now it seems like kind ofwe're backtracking and it's making
harder to discover actions andwhat you need to do in an app.
And so that's what I've been trying tofigure out as well is like for one example
is the tab bar minimization, right?
Like, that seems like something thatthey're recommending, for a app to
(16:02):
do is like minimize the tab bar.
But I remember like in theguidelines around tab bars is
like they explicitly said tab barshould always be present, right?
have they updated theguidelines at all since June?
yes, they have updated HIG thehuman interface guidelines.
they have updated them, but when youwatch some of the old dub videos and
(16:25):
now like you think about the new things.
I'm not understanding some of thedecisions that were made or like why some
of these design changes have been made.
And so, yeah, I decided, I'm not goingto do the minimize tab bar thing.
it doesn't seem like,a good user experience.
I think that's been the hardestthing is just like, it seems
like it's a movie target.
'cause things are changing so muchin the betas and it's not always
(16:47):
clear the best way to, integratefor the best user experience
Right.
what's the frosted opacity in this beta?
except
What other features were you interestedin taking advantage of, with iOS 26.
The other two, that have been,that I've really been excited about
is alarm kit and speech analyzer.
(17:09):
Yes.
Yeah.
I was like, so when I saw that inthe notes alarm kit, I was like,
finally, okay, now I understand.
Like for me, whenever I wanted to dosomething like that, I would just post
notifications and go like that, right?
okay, now I understand whywe need alarm kit as opposed
to just using notifications.
I'll let you explain why you're usingAlarm Kit and what you like about it.
(17:33):
I think obviously our last conversationwas, a lot about notifications, right?
So
That's true.
yeah, so obviously I'm an advocatefor notifications, but I think for
some apps, notifications are overkill.
And I think notifications are notnecessarily, the best thing for
every app because notificationsare easily dismissed and forgotten.
(17:54):
Right?
yeah, so I am very disappointed inthe use of notifications nowadays
'cause it's so confusing how it's beenmisused and abused by like marketing.
Well, at the same time it's some things,it's I want be notified because of this.
I don't need to be notified becauseyou have a new weekly ad for whatever.
And that's been really abused.
Unfortunately.
(18:15):
That's made it really hard to not use itas a user using notifications, because
like sometimes I miss stuff now becausethere's so many notifications and it's
yeah, I just went through like a auditof all my apps to turn off notifications.
'cause I was getting sofrustrated with missing things, so
modes don't help.
Unfortunately.
I wish I understoodthose better, but yeah.
(18:38):
Yeah.
So I feel like Alarm Kit is gonnabe great for developers because
for apps where you want to.
The user to take an immediate action, orthey want to be like notified in a way
where it's okay, this is like an immediatething you need to take action on.
It's a great way to do that.
And it's much simpler to implementthe notifications, I think.
(18:59):
Like I said, notifications, you know,it can be a lot to, to implement.
From the developer's standpoint, it isa lot to think about and manage when you
think about the notification center andeverything and whether you're doing server
side notifications, like there, there'sall these things to think about, but
Alarm kit so far is fairly simple, right?
It comes with its own manager scheduling.
Canceling, updating Notifica.
(19:20):
The alarms are super simple.
And so I think it's gonna be a great wayfor apps that don't need notifications,
that just want the user to takean immediate action in their app.
A way to do that really simply.
I do have a question beforewe get to speech analyzer.
what are you using to storestuff locally and do you have
(19:41):
a server side component to it?
Yeah, so I don't have aservice side component.
I am using a sharing GRDB for
I laugh because, I'll ask the question,why didn't you not go with swift data?
So I actually started out with swift data.
yeah, I think I saw atweet that you said that,
(20:01):
so I started out with swift data, andit just was, it's not robust enough.
The predicate handling, is still lackingin terms of more like complex predicates.
The cloud kit support onlysupports, private, cloud kit,
databases doesn't support shared.
And so for me, having the abilityfor parents to be able to share
(20:24):
data is like essential, right?
that's where I was going with the serverside question is, if Dad is gonna feed the
baby with formula or mom's pumped milk,like it'd be nice to have that component.
Exactly.
So to be able to supportcloud kit was essential.
I also felt like swift data.
I really like, you know, using structsand, and so having, being forced to have
(20:47):
my models all be classes, yeah, so I,I, I think that there's just like a some
limitations that with swift data whereI started out with it and I was a bit
excited, but then as I started to, tostart doing things that were a little bit
more complex, I just ran into a lot ofthe limitations and so decided to move on
point free code is doing a lot ofgreat work with, their libraries,
(21:08):
their open source libraries, andso the sharing GRDB has been great.
And they're working on a cloudkit, library as well to support it.
And so, yeah, I've had a great experienceimplementing my database layer that way.
I have, yeah, I've started being moreand more disappointed and swift out on
the headaches I've run into with it.
So I support your decision.
(21:29):
I get it.
Are you gonna look at some sort of server?
what are you thinking?
Cloud Kit or are youthinking like Vapor or what?
Yeah, I'm definitely doing cloud kit.
I haven't really thoughtabout server side, but
I mean, cloud Kit would do
Right?
it does everything.
It's free.
Yeah.
And I don't like, have to deal with,authentication and account creation
(21:51):
and everything, which is nice.
But, you know, I have thought ifin the future, you know, folks, if
there's enough interest and folks areexpressing interest on the Android side
of things, I've thought about that.
'cause like with.
My, RI two, I did end up doingAndroid version, but, so I have
(22:13):
thought about that, like, okay, inthe future will I need to have like a
server side component and think aboutthat, but I'm not starting there.
I'll deal with that downthe line if there's a need.
I agree with that.
You got bigger fish, the fry.
Yeah, totally.
Yeah.
Cool.
Okay, so let's do it.
Speech analyzer.
So you had talked earlier about how.
(22:34):
It's really hard for mom totap on the phone and do stuff.
I've also, grilled food and havehad notifications on my watch
while I'm carrying something.
So using my nose is not mypreference for interacting with
my watch, so I totally get it.
So I'm assuming that like speech analyzeris the same idea where it's like, oh,
(22:56):
I don't have to like touch the thing.
I could just say, Hey, Dingus.
I pumped three ounces of milkright now, or something like that.
Is that, am I saying that correctly?
Yeah.
Well, I mean there, there is.
Still the, you have toinitiate via touch, right?
Like the recording.
But at least like, there'sless, I think, you know,
(23:17):
interaction that's, that's needed.
So there's Speech analyzer hasnot related to what I think
I was assuming was like Siri
Yeah.
Maybe explain that I
Yeah, so speech analyzer is the modelthat is doing the transcription.
And so it's a new on device model that is.
Better than the oldlike SF Speech Analyzer.
(23:38):
It's much faster.
Doesn't require the server dependency.
It manages like the locales and language,the languages for you, et cetera.
So it's the transcription side of things.
Now, I haven't dived into Siri Kit,which is now I think activity kit.
I think, yeah, there's all that.
I wanna dive
(23:58):
Or intense
Exact app.
Intense.
Yeah, I haven't dived into that yet,which, but I will have to, 'cause
I wanna support live activity,I wanna have widgets, et cetera.
So I definitely willhave to dive in there.
And I haven't seen if it'spossible to support, transcription.
Inside of an app intent.
I think that would be really powerful.
(24:19):
So like if I could have it bewhere it's like, Hey, Siri, you
know, login, milk diary, et cetera.
If there's a way to, to supportthat, I think that would be awesome.
Awesome.
'cause that would be completelyhands off, like you said.
but I haven't dived into that yet.
So it sounds to me like speechanalyzer is an in-app experience
that gives you a lot more powerthan what Siri would be able to do
(24:41):
Yeah.
I think the cool thing is like.
With the foundation models, right?
It's all prompt based.
So the speech analyzer being a bettermodel for audio transcription allows you
to then just take that audio and thenfeed it directly into a prompt, right?
And so I think it justallows more powerful features
(25:02):
and integrations that way.
Yep.
Okay.
Very cool.
I wanna definitely wannatake a look at this.
This sounds awesome.
anything else you wanted tomention before we close out?
I mean, I think the main thing I'vebeen thinking, you know, technology
and everything is changing a lot.
Like this is a very, I think,interesting time to be a developer.
(25:24):
but I think the exciting thing, and one ofthe reasons why I got into app development
in the first place is the ability to make.
Cool technology thathelps people in some way.
And so I think when I think about, folks,I hope that people still feel inspired
and like have the energy and enthusiasm,to still wanna make technology that
(25:47):
helps people among all of the talk ofiOS 26, and I know, there's a lot of
controversy and upset, but I still feelreally excited about Apple platforms
and being an Apple platform developer.
did you use any AI toolsfor building your app
I have, used Claude code
Yep.
I have used Claude code.
I think Rob Napier had a really good,blog post about, LLMs and how they're
(26:11):
basically like really good interns.
And I think that when you think aboutit from that perspective, it's not like
a replacing your development and codingexperience, but it's, a support of, and
I think as a solo developer, when youdon't have a team, and folks to be able
to bounce ideas off of or get feedback,I think that's what I found it really
(26:35):
helpful for is like, okay, review thiscode that I wrote, like where are the
downfalls, like how can I improve it?
a tool I just signed up foron GitHub, called Code Rabbit.
It's a AI tool for doingpull request reviews.
that's kind of like what you're
Yeah, exactly, because I just feellike it, especially when you're a solo
(26:55):
developer, it can be hard sometimes,like, it's easy to just, write your
code and think everything is good, butyou don't have that kind of feedback
that you get that's really great whenyou're working in a team environment.
So I found it really helpful forthat and just, not having it right.
Everything for me or anything, butjust more, trying to improve what I
am doing and improve my development,and make things a bit faster if I
(27:18):
need help, writing something quickly.
It's been really helpful for test.
'cause that's one thing that, youknow, I test driven development
Do you have, Claudewrite unit tests for you?
yeah, I have been doing that,
That's really good.
To have it right, test and make sureI'm covering all my bases there.
So yeah, I have found it helpfulin that way to be more like a pair
(27:39):
programmer, intern type of experience.
How about stuff outside ofaccounting, like marketing
or copy or any of that stuff?
I think, what I usually do is.
I just ask it for a bunch of,like, variations on things
that I already have written.
So I always start, human first.
I'll write things and just saylike, okay, can you gimme some
(28:00):
variations of this, in my writingstyle, what would you recommend?
How can I make this a littlebit more concise and whatnot?
And so I have found it helpful for that.
When you're a solo team, it isjust a helpful tool give feedback
and be able to brainstorm.
I found it really helpful in that way.
I have a lot of thoughts on AI and coding.
Yeah.
so I am working on a series ofblog posts and a video on it,
(28:23):
Oh, okay.
Yeah, I'm excited to, check that out.
So are you like a proponent or.
I'm not gonna say,
Okay,
I use it a lot.
So I'm not a. there's such anextreme spectrum it feels like,
where people are like, AI is great.
I wanted to take over the worldand destroy the human race.
And it's wonderful.
(28:45):
And then there's the, youknow, what do you call it?
The people who are like, now AI is awful.
It's garbage.
Like we should stop usingit and blah, blah, blah.
I would say I'm in the middle, buteverybody kind of would say they're
in the middle in anything, so I haveto like, not sure where I'd be, but.
It's interesting you mention feedbackbecause I do think one of its weaknesses,
(29:10):
and you see this in a lot of a lotof places, is it will always tell
you you're right and how great youare, and so that's my one weakness.
I've found amongst others, like howwell they're trained, if they're
trained on new stuff, because youreally don't wanna keep using UI kit
in 2025 and still trained on that.
(29:32):
That is such a pain.
as a Mac developer, I run into that a lot.
There's like all sorts of new Ns.
Classes that don't exist that it createsbecause it sees the UI kit version.
I do think one of its problemsis it'll tell you're right.
So it's good that you're usingit for like code reviews.
'cause I do think that's super helpful.
(29:52):
And other problem ispeople get lazy, right?
That's kind of the core of it islike AI replaces work, so then you
just get lazy and assume it's doingthe right thing, which it's not
It is not.
Yeah, I think that's the number one thingI've learned with it is like I never
assume it's right and I never assume thatanything that it gives me is correct.
So you always I think people talk about,oh yeah, you're saving a lot of time.
(30:14):
And to be frank, I don't know how muchtime I save with it because I do a lot
Error proofing.
Right.
Because it does a lot of hallucinationsand so, I do spend a lot of
time like correcting it and likereminding it of what the correct
APIs are and things like that.
it does something wrong, I'vebeen trapped in rabbit holes
where it's like I'm spending waytoo much time trying to tell it.
(30:34):
No, you're doing it wrong.
Yeah.
So
Kay, it was good to see you.
I'm looking forward to havingyou back on in five years.
This is great
maybe we could do itearlier than five years.
Next time.
We can learn about your newapp for helping kindergartners.
Thank you so much.
This is awesome.
I'm so happy to see you again.
where can people find you online?
(30:55):
Where can people find milk?
Diary?
Is it available?
Test, flight, or, signup or Yeah, go ahead.
Plug your, plug yourself away.
Okay.
Yeah, so milk diary you canfind, at milk diary.com.
the test flight is not opento the public just yet.
I'm still in private alpha,but I will be open up.
Slowly, testing opportunities.
(31:17):
you can sign up for the mailinglist on the website and then you
can find me online on Mastodon.
Kaya Thomas is, and blue sky.
Although to be honest, I, I'm not reallyon blue sky that much, or threads.
I'm also on threads.
Same thing, Kaya Thomas is.
my website is also Kaya Thomas is so youcan find me online, all those places.
(31:41):
Kaya.
Thank you so much.
People can find me online at Leo Dion.
At Leo Dion at CI am on Mastodon.
my company is bright Digit.
We have all sorts of new stuff on there.
I just posted the iOS backend cheatsheet, so if you're looking for
what kind of backend to pick, youmight want to take a look at that.
(32:03):
And yeah, it was great foreverybody to join and thank you
so much and have a great summer.
Talk to you later.
Bye.
Welcome to another episode, empower Apps.
I'm your host, Leo Dion.
Today I am joined by Kaya ThomasThank you for coming on the show.
Of course, Leo, I'm glad to be here.
Thanks for having me again.
So what have you been upto since, five years ago.
Pre COVID.
Yeah.
What have you been up to?
(00:24):
Oh man.
it's been a lot in the last five years.
I, took a break from iOS developmentand went outside my comfort zone and
went to grad school and, decided todo some more customer facing work and.
Work on the people side of things.
And, it's been a whirlwind and agreat experience, I think to step
outside of engineering a bit.
(00:46):
And now I'm, trying to mixboth worlds and make a new app.
That's awesome.
So I actually looked it up.
Our episode was like rightat the beginning of COVID,
just to give you an idea.
It was March the 2020, which is hilarious.
The start of it all.
right.
That's when we thought KobeBryant passing away was gonna
be the worst thing of that year.
(01:06):
So Yeah.
It got a lot worse.
So.
Let's talk about thisbrand new app, milk Diary.
You wanna talk about why did you makethis and what did you find, what do you,
what did you find was the need that youfeel like Milk Diary is gonna help with?
Yeah, so I became a mom last year andyou know, that's a huge life change.
(01:31):
And of course, because I love apps.
I wanted to use apps.
As a part of my parenting experienceto help manage and track things.
And so one of the most intensiveparts of parenting, a newborn and
a baby is feeding them and the,
Mental, logistical things you have tokeep track of, and the physicality of it.
(01:56):
But there's a lot to think about,like how often they need to feed.
How much milk do you have for them?
access to them to access, like howmuch you have in your fridge, how
much do you have in your freezer?
Just understanding the differencebetween you feeding them and
you actually making the milk.
So there's a lot of like logisticaland administrative stuff.
(02:17):
And obviously, you know, as developerswe know that the, the great thing about
technology is can help with some of likethe logistical and administrative stuff.
But I was getting frustrated 'cause a lotof the apps, we're not flexible enough and
we're not, really thinking about this kindof full experience and all the different
(02:38):
connections between how you have to makethe milk track, the milk, store, the milk.
Like it's, it's just likethere's this whole holistic.
Journey and process there.
And I felt like all the apps thatI were using might do one thing,
but then when you need somethingelse, you can't do it in that app.
So you might have to use anotherapp or track it another way.
(02:59):
I started to get the itch again andI was like, okay, I think I need to
create an app because, I felt likethere needs to be something better.
And I think that, especially as a mom,I realized that a lot of the apps out
there are not actually made by folks whoare going through this experience, right.
And living through it.
And so having that.
(03:19):
Experience and being able to also callupon, moms, my friends and just moms in
the community and get their experience aswell and be able to really make something
that's catered to the experience ofnew parents and understand what they
would need and how to make an app thatis, user friendly, easy to use, and
actually like saving time and energy.
(03:42):
Well, what does yourdaughter think of the app?
I mean, honestly, She is, I think enamoredwith, technology in general just because
the idea of just touching the screen andhaving something happen, that reaction.
So I don't think she really understandslike, what an app is per se, but
she loves pressing things and,
Ima, yes, I
(04:02):
and seeing it react on the screen.
Yes.
Yeah.
so I asked Mrs. Dion beforerecording, what did she do?
And she just used the notes app.
So, which I'm not surprised.
Or a spreadsheet 'causeshe's loves spreadsheets.
this looks really awesome.
what.
One of the things that you talked abouthere was that really piqued my interest
(04:25):
was the use of foundation models.
Yeah.
what were you thinking as far ashow that would fit in with your app?
Yeah.
So before, dub I was reallystruggling with the idea of like
smart intelligence features, right?
Because I figured thatintegrating some type of.
(04:48):
Intelligence would be great for the appbecause it could help with that more
like efficiency and kind of logisticalthing that I was talking about.
But I was really hesitant to integrateany of the cloud models, like any
of the LLMs that are existing.
Open, open AI and anthropic, et cetera.
Because this data's really sensitive,like I don't think a lot of parents want
(05:12):
this type of data, being sent right overthe internet or being sent somewhere.
And so I was reallyhesitant to integrate that.
And so when they announced foundationmodels, I was like, yes, this
is my opportunity to be able to,incorporate intelligent features
in a really private way on device.
So it was just a huge win.
(05:34):
I think the ways in which I'mintegrating it is a couple ways.
So one is being able to logsessions via text or voice, right?
So all of the apps where you track thingsaround, you have to manually enter right?
Oh, here's how much, and you haveto go to the different text inputs
and, select the different things.
(05:56):
And a lot of times, as anew parent, you're tired.
you don't wanna be like, doing a
have an extra hand 'causeyou're carrying someone.
Exactly.
So being able to just either,you know, input text, or even
speak right into the app and say,okay, like I pumped five ounces.
And then just have it save.
So just being able to save that time.
So that's where the foundation modelsreally come in to be able to have
(06:18):
those prompts that allow them to createthe data, that they want to track.
And I think the difference I foundwith the foundation models really
helped me from a developer side.
I was really overwhelmed.
When I started to think about thecloud models and like how I would
integrate them in a way that wouldreduce hallucinations, get actual data
(06:39):
that, was focused right and accurate.
So the fact that the foundation modelshave guided generation is just so good.
I mean, I think that.
You have to work on tweaking your datamodels to make sure that they work.
So you still reduce hallucinations'cause that's just a thing with LLMs.
(07:00):
But it's much better.
And I think as a developer, it allowsme to have a bit more reliability
and consistency on what I'm gettingfrom the prompts, the user prompts.
So we're using,
I mute Gen General.
General
General, is it a macro or a protocol?
Okay, so you're using that and thenthat would be the output that you're
(07:24):
looking for when somebody sayssomething and that way it's actionable.
Correct.
Yeah, so how I'm doing it and ittook a while for me to figure out
the best way to think about thedata schema for the general structs.
At first I was thinking I would just havea session struct and, you know, it would
have all the data, whether it was a pumpsession, nursing, a bottle feed, right?
(07:48):
And the foundation model shouldbe able to detect, based off what
they're saying, what type of.
Feed it is and be ableto fill in the data.
But I realized actually, at leastin my experience with working with
the foundation model so far, is it'sactually better to have smaller,
structs and have them be very specific.
So breaking them out, becauseI realized that when you have
(08:11):
just like a kind of catchall.
data model for the foundations, thehallucinations increase and it's a bit
harder for it to figure out, exactlywhich variables it should fill in
based off of the context of the prompt.
So you're using like enums and stuffto make it as specific as possible?
Yeah.
Rather than just like a stringwhere you can get something crazy.
(08:33):
so there's both the feedingand the pumping, right?
Are there things you'retracking in the app
Yeah, so it's bottle feeding,nursing, and pumping.
Okay.
And then you also mentioned formula.
how does that fit in?
what's the use case for that?
Yep.
A lot of parents combo feed, right?
So they use breast milk and formulain the app you'll have the ability
(08:55):
to track formula or breast milk.
if you're doing bottle feeds withformula, you'll be able to track
what type of formula you use.
I know a lot of parents also try differenttypes of formula, you can even keep track
of notes to see if there was reactions or
Got it.
things like that.
youngest daughter, so I don't knowif you know my personal story,
but our four oldest are, adopted.
(09:15):
at least the youngest, I think theyoungest girl who's now gonna be 11, she
moved in with us when she was a newbornand we had a lot of formula I remember
doing EXCO development and having her onmy lap while I'm feeding her at 2:00 AM
Oh wow.
I totally remember that.
But.
Formula.
doesn't last as long, right?
(09:37):
As breast milk, as far as I'm concerned.
Yeah.
So, yeah, I could totallysee how that would fit in.
And I also, like one thing you mentionedhere, is, twins or multiples, right?
like having to track multiple children.
That's a really important niche
Yes.
And I have, a close friend who hastwins and I realized in testing out.
(09:57):
A lot of the apps is all of themmake you switch between profiles
in order to track anything.
So you have to create, a profile foreach kid and then if you wanna track
anything, you have to switch, likebetween profiles, to track a feed or.
Yes.
when I thought about like, okay,how am I gonna model this, app?
(10:18):
I was like, okay, from the start,I wanna make it super easy if you
have multiples and you don't have toswitch profiles or do anything, extra.
If you want to just track something andyou just select which kid it's for, right?
So this is totally iOS, 26, correct.
it
iPhone only.
iPhone only.
iPhone only to start.
(10:38):
I think that, if I find that there'sinterest in iPad support or Mac
support, I would think about it.
But yeah, iPhone only
watch support.
Oh yeah, watch support is definitelysomething I want to have because
that is something I used a lotactually when I was feeding.
Not vision os I don't see
yeah,
really useful.
I wanna ask this question, butdid you do like a comparative
(10:59):
analysis before you got started?
Just to see what was currently out there?
And what was themarketing angle for those?
How do they monetize?
Yeah.
Are they basically crappy React apps?
what's the environment out there?
Yeah, it's interesting.
I think so I definitely did a comparativeanalysis and there are, like, to be
(11:19):
frank, a lot of apps out there that,track baby, newborn tracking and,
feed tracking and stuff like that.
But I think, there's definitely,you have the cross platform, popular
cross platform apps that are, in myopinion, I, despise using, because,
I like native apps and I find thatthey were a poor user experience.
(11:41):
I think what I found is.
You have a lot of apps that aretrying to do everything, so they're
like, oh, you can track everything.
Right, but
the thing I was thinking, sorry.
It was just like, oh, Gerberhas a app and it's you know,
it's their app for everything.
And then by the way, we have this littlemilk diary thing over here in the corner.
That's kind of what I was thinkingwas gonna end up happening.
(12:03):
Yeah, exactly.
And then it is the user experienceis super clunky and like, it's
hard to use and know how you eventrack things and where you do it.
I just found it's very overwhelming.
A lot of these apps because they'retrying to do too many things.
And so for me, that's one of thedecisions I made is I really wanted to
focus on the full feeding experienceand be able to support everything
(12:24):
in that space and not try to be thiscatchall that's doing everything.
Because I think that oftentimeswhat makes apps great is when
there is a specific focus, right?
And you're getting really, thevalue out of it because it.
Can help you solve problemsfor a specific need.
a lot of the apps out there, there'ssome free apps out there that have a
(12:44):
bunch of ads, they monetize via ads.
then there's some that are trying tolean into, I think a big area that
some of the larger apps are trying tolean into is sleep tracking and sleep
predictions Personally I don't think that.
Having experienced the ups and downsof a newborn and infant sleep, I
don't think an app can really predict,
(13:05):
Or even like, what'sthe point in tracking?
It's not like you can be like, all right,sweetie, but you can't do the getting up
at 2:00 AM This is what my watch said.
exactly.
I don't, there's nothing actionable.
Exactly.
So like, I mean, there's a bunchthat are leaning into that.
And then like you said, a lot ofthe formula companies have apps
and stuff that I think are, a meansto sell their formula and stuff.
(13:25):
And so, there are a couple of, indieapps that are nice, but I think because.
They are not, like I said, they'revery specific and they don't support.
So they might support one thing reallywell, but they're not thinking about the
full journey of the feeding experience.
And so it's not flexible enough, ifsomething changed about your journey.
(13:46):
So, that's where I feel like I can fillthat space and that gap, to be able to
provide something that's comprehensiveand flexible enough to use throughout
that entire first year from newborn to.
early toddler days,
So when you first developed theapp, were you developing it for iOS?
18.
I was,
Okay.
So.
(14:09):
I wanna really take advantageof everything new, including
foundation models, but also yougot liquid glass to deal with.
How has that gone?
It,
oh
li liquid glass, it's a moving targetbecause every beta, right, like there's
something a little bit different.
I think the most difficult part for mehas been understanding like the best way
(14:35):
to embrace it and where to embrace it.
Do you even have a choice?
for something like tab bars, you don'thave a choice, but when you think about
like buttons and toolbar items and things,how you integrate it, you have a choice in
terms of how it looks and how far you go.
Especially with custom UI elements,How much you integrate, liquid glass.
(14:56):
So I think that's been the hard part offiguring out it's not necessary to have
it for every button And every interaction.
where do you prioritize having it and whatis the prioritization for liquid glass?
Like is it just for primary actions?
It's, so, I think that's whatI've been trying to figure out.
I think the thing that I really don'tlike, about it is, it seems as if
(15:18):
when I look at like the Apple appson, in the beta, right, like notes and
reminders and mail, et cetera, it seemsthat there's a lot of hiding actions
behind menus and, and things thatwhere before at least my understanding
of like the design guidelines.
What's recommended is you wanna makeit easier for folks to know where
(15:41):
to go and what actions to take.
Right.
And now it seems like kind ofwe're backtracking and it's making
harder to discover actions andwhat you need to do in an app.
And so that's what I've been trying tofigure out as well is like for one example
is the tab bar minimization, right?
Like, that seems like something thatthey're recommending, for a app to
(16:02):
do is like minimize the tab bar.
But I remember like in theguidelines around tab bars is
like they explicitly said tab barshould always be present, right?
have they updated theguidelines at all since June?
yes, they have updated HIG thehuman interface guidelines.
they have updated them, but when youwatch some of the old dub videos and
(16:25):
now like you think about the new things.
I'm not understanding some of thedecisions that were made or like why some
of these design changes have been made.
And so, yeah, I decided, I'm not goingto do the minimize tab bar thing.
it doesn't seem like,a good user experience.
I think that's been the hardestthing is just like, it seems
like it's a movie target.
'cause things are changing so muchin the betas and it's not always
(16:47):
clear the best way to, integratefor the best user experience
Right.
what's the frosted opacity in this beta?
except
What other features were you interestedin taking advantage of, with iOS 26.
The other two, that have been,that I've really been excited about
is alarm kit and speech analyzer.
(17:09):
Yes.
Yeah.
I was like, so when I saw that inthe notes alarm kit, I was like,
finally, okay, now I understand.
Like for me, whenever I wanted to dosomething like that, I would just post
notifications and go like that, right?
okay, now I understand whywe need alarm kit as opposed
to just using notifications.
I'll let you explain why you're usingAlarm Kit and what you like about it.
(17:33):
I think obviously our last conversationwas, a lot about notifications, right?
So
That's true.
yeah, so obviously I'm an advocatefor notifications, but I think for
some apps, notifications are overkill.
And I think notifications are notnecessarily, the best thing for
every app because notificationsare easily dismissed and forgotten.
(17:54):
Right?
yeah, so I am very disappointed inthe use of notifications nowadays
'cause it's so confusing how it's beenmisused and abused by like marketing.
Well, at the same time it's some things,it's I want be notified because of this.
I don't need to be notified becauseyou have a new weekly ad for whatever.
And that's been really abused.
Unfortunately.
(18:15):
That's made it really hard to not use itas a user using notifications, because
like sometimes I miss stuff now becausethere's so many notifications and it's
yeah, I just went through like a auditof all my apps to turn off notifications.
'cause I was getting sofrustrated with missing things, so
modes don't help.
Unfortunately.
I wish I understoodthose better, but yeah.
(18:38):
Yeah.
So I feel like Alarm Kit is gonnabe great for developers because
for apps where you want to.
The user to take an immediate action, orthey want to be like notified in a way
where it's okay, this is like an immediatething you need to take action on.
It's a great way to do that.
And it's much simpler to implementthe notifications, I think.
(18:59):
Like I said, notifications, you know,it can be a lot to, to implement.
From the developer's standpoint, it isa lot to think about and manage when you
think about the notification center andeverything and whether you're doing server
side notifications, like there, there'sall these things to think about, but
Alarm kit so far is fairly simple, right?
It comes with its own manager scheduling.
Canceling, updating Notifica.
(19:20):
The alarms are super simple.
And so I think it's gonna be a great wayfor apps that don't need notifications,
that just want the user to takean immediate action in their app.
A way to do that really simply.
I do have a question beforewe get to speech analyzer.
what are you using to storestuff locally and do you have
(19:41):
a server side component to it?
Yeah, so I don't have aservice side component.
I am using a sharing GRDB for
I laugh because, I'll ask the question,why didn't you not go with swift data?
So I actually started out with swift data.
yeah, I think I saw atweet that you said that,
(20:01):
so I started out with swift data, andit just was, it's not robust enough.
The predicate handling, is still lackingin terms of more like complex predicates.
The cloud kit support onlysupports, private, cloud kit,
databases doesn't support shared.
And so for me, having the abilityfor parents to be able to share
(20:24):
data is like essential, right?
that's where I was going with the serverside question is, if Dad is gonna feed the
baby with formula or mom's pumped milk,like it'd be nice to have that component.
Exactly.
So to be able to supportcloud kit was essential.
I also felt like swift data.
I really like, you know, using structsand, and so having, being forced to have
(20:47):
my models all be classes, yeah, so I,I, I think that there's just like a some
limitations that with swift data whereI started out with it and I was a bit
excited, but then as I started to, tostart doing things that were a little bit
more complex, I just ran into a lot ofthe limitations and so decided to move on
point free code is doing a lot ofgreat work with, their libraries,
(21:08):
their open source libraries, andso the sharing GRDB has been great.
And they're working on a cloudkit, library as well to support it.
And so, yeah, I've had a great experienceimplementing my database layer that way.
I have, yeah, I've started being moreand more disappointed and swift out on
the headaches I've run into with it.
So I support your decision.
(21:29):
I get it.
Are you gonna look at some sort of server?
what are you thinking?
Cloud Kit or are youthinking like Vapor or what?
Yeah, I'm definitely doing cloud kit.
I haven't really thoughtabout server side, but
I mean, cloud Kit would do
Right?
it does everything.
It's free.
Yeah.
And I don't like, have to deal with,authentication and account creation
(21:51):
and everything, which is nice.
But, you know, I have thought ifin the future, you know, folks, if
there's enough interest and folks areexpressing interest on the Android side
of things, I've thought about that.
'cause like with.
My, RI two, I did end up doingAndroid version, but, so I have
(22:13):
thought about that, like, okay, inthe future will I need to have like a
server side component and think aboutthat, but I'm not starting there.
I'll deal with that downthe line if there's a need.
I agree with that.
You got bigger fish, the fry.
Yeah, totally.
Yeah.
Cool.
Okay, so let's do it.
Speech analyzer.
So you had talked earlier about how.
(22:34):
It's really hard for mom totap on the phone and do stuff.
I've also, grilled food and havehad notifications on my watch
while I'm carrying something.
So using my nose is not mypreference for interacting with
my watch, so I totally get it.
So I'm assuming that like speech analyzeris the same idea where it's like, oh,
(22:56):
I don't have to like touch the thing.
I could just say, Hey, Dingus.
I pumped three ounces of milkright now, or something like that.
Is that, am I saying that correctly?
Yeah.
Well, I mean there, there is.
Still the, you have toinitiate via touch, right?
Like the recording.
But at least like, there'sless, I think, you know,
(23:17):
interaction that's, that's needed.
So there's Speech analyzer hasnot related to what I think
I was assuming was like Siri
Yeah.
Maybe explain that I
Yeah, so speech analyzer is the modelthat is doing the transcription.
And so it's a new on device model that is.
Better than the oldlike SF Speech Analyzer.
(23:38):
It's much faster.
Doesn't require the server dependency.
It manages like the locales and language,the languages for you, et cetera.
So it's the transcription side of things.
Now, I haven't dived into Siri Kit,which is now I think activity kit.
I think, yeah, there's all that.
I wanna dive
(23:58):
Or intense
Exact app.
Intense.
Yeah, I haven't dived into that yet,which, but I will have to, 'cause
I wanna support live activity,I wanna have widgets, et cetera.
So I definitely willhave to dive in there.
And I haven't seen if it'spossible to support, transcription.
Inside of an app intent.
I think that would be really powerful.
(24:19):
So like if I could have it bewhere it's like, Hey, Siri, you
know, login, milk diary, et cetera.
If there's a way to, to supportthat, I think that would be awesome.
Awesome.
'cause that would be completelyhands off, like you said.
but I haven't dived into that yet.
So it sounds to me like speechanalyzer is an in-app experience
that gives you a lot more powerthan what Siri would be able to do
(24:41):
Yeah.
I think the cool thing is like.
With the foundation models, right?
It's all prompt based.
So the speech analyzer being a bettermodel for audio transcription allows you
to then just take that audio and thenfeed it directly into a prompt, right?
And so I think it justallows more powerful features
(25:02):
and integrations that way.
Yep.
Okay.
Very cool.
I wanna definitely wannatake a look at this.
This sounds awesome.
anything else you wanted tomention before we close out?
I mean, I think the main thing I'vebeen thinking, you know, technology
and everything is changing a lot.
Like this is a very, I think,interesting time to be a developer.
(25:24):
but I think the exciting thing, and one ofthe reasons why I got into app development
in the first place is the ability to make.
Cool technology thathelps people in some way.
And so I think when I think about, folks,I hope that people still feel inspired
and like have the energy and enthusiasm,to still wanna make technology that
(25:47):
helps people among all of the talk ofiOS 26, and I know, there's a lot of
controversy and upset, but I still feelreally excited about Apple platforms
and being an Apple platform developer.
did you use any AI toolsfor building your app
I have, used Claude code
Yep.
I have used Claude code.
I think Rob Napier had a really good,blog post about, LLMs and how they're
(26:11):
basically like really good interns.
And I think that when you think aboutit from that perspective, it's not like
a replacing your development and codingexperience, but it's, a support of, and
I think as a solo developer, when youdon't have a team, and folks to be able
to bounce ideas off of or get feedback,I think that's what I found it really
(26:35):
helpful for is like, okay, review thiscode that I wrote, like where are the
downfalls, like how can I improve it?
a tool I just signed up foron GitHub, called Code Rabbit.
It's a AI tool for doingpull request reviews.
that's kind of like what you're
Yeah, exactly, because I just feellike it, especially when you're a solo
(26:55):
developer, it can be hard sometimes,like, it's easy to just, write your
code and think everything is good, butyou don't have that kind of feedback
that you get that's really great whenyou're working in a team environment.
So I found it really helpful forthat and just, not having it right.
Everything for me or anything, butjust more, trying to improve what I
am doing and improve my development,and make things a bit faster if I
(27:18):
need help, writing something quickly.
It's been really helpful for test.
'cause that's one thing that, youknow, I test driven development
Do you have, Claudewrite unit tests for you?
yeah, I have been doing that,
That's really good.
To have it right, test and make sureI'm covering all my bases there.
So yeah, I have found it helpfulin that way to be more like a pair
(27:39):
programmer, intern type of experience.
How about stuff outside ofaccounting, like marketing
or copy or any of that stuff?
I think, what I usually do is.
I just ask it for a bunch of,like, variations on things
that I already have written.
So I always start, human first.
I'll write things and just saylike, okay, can you gimme some
(28:00):
variations of this, in my writingstyle, what would you recommend?
How can I make this a littlebit more concise and whatnot?
And so I have found it helpful for that.
When you're a solo team, it isjust a helpful tool give feedback
and be able to brainstorm.
I found it really helpful in that way.
I have a lot of thoughts on AI and coding.
Yeah.
so I am working on a series ofblog posts and a video on it,
(28:23):
Oh, okay.
Yeah, I'm excited to, check that out.
So are you like a proponent or.
I'm not gonna say,
Okay,
I use it a lot.
So I'm not a. there's such anextreme spectrum it feels like,
where people are like, AI is great.
I wanted to take over the worldand destroy the human race.
And it's wonderful.
(28:45):
And then there's the, youknow, what do you call it?
The people who are like, now AI is awful.
It's garbage.
Like we should stop usingit and blah, blah, blah.
I would say I'm in the middle, buteverybody kind of would say they're
in the middle in anything, so I haveto like, not sure where I'd be, but.
It's interesting you mention feedbackbecause I do think one of its weaknesses,
(29:10):
and you see this in a lot of a lotof places, is it will always tell
you you're right and how great youare, and so that's my one weakness.
I've found amongst others, like howwell they're trained, if they're
trained on new stuff, because youreally don't wanna keep using UI kit
in 2025 and still trained on that.
(29:32):
That is such a pain.
as a Mac developer, I run into that a lot.
There's like all sorts of new Ns.
Classes that don't exist that it createsbecause it sees the UI kit version.
I do think one of its problemsis it'll tell you're right.
So it's good that you're usingit for like code reviews.
'cause I do think that's super helpful.
(29:52):
And other problem ispeople get lazy, right?
That's kind of the core of it islike AI replaces work, so then you
just get lazy and assume it's doingthe right thing, which it's not
It is not.
Yeah, I think that's the number one thingI've learned with it is like I never
assume it's right and I never assume thatanything that it gives me is correct.
So you always I think people talk about,oh yeah, you're saving a lot of time.
(30:14):
And to be frank, I don't know how muchtime I save with it because I do a lot
Error proofing.
Right.
Because it does a lot of hallucinationsand so, I do spend a lot of
time like correcting it and likereminding it of what the correct
APIs are and things like that.
it does something wrong, I'vebeen trapped in rabbit holes
where it's like I'm spending waytoo much time trying to tell it.
(30:34):
No, you're doing it wrong.
Yeah.
So
Kay, it was good to see you.
I'm looking forward to havingyou back on in five years.
This is great
maybe we could do itearlier than five years.
Next time.
We can learn about your newapp for helping kindergartners.
Thank you so much.
This is awesome.
I'm so happy to see you again.
where can people find you online?
(30:55):
Where can people find milk?
Diary?
Is it available?
Test, flight, or, signup or Yeah, go ahead.
Plug your, plug yourself away.
Okay.
Yeah, so milk diary you canfind, at milk diary.com.
the test flight is not opento the public just yet.
I'm still in private alpha,but I will be open up.
Slowly, testing opportunities.
(31:17):
you can sign up for the mailinglist on the website and then you
can find me online on Mastodon.
Kaya Thomas is, and blue sky.
Although to be honest, I, I'm not reallyon blue sky that much, or threads.
I'm also on threads.
Same thing, Kaya Thomas is.
my website is also Kaya Thomas is so youcan find me online, all those places.
(31:41):
Kaya.
Thank you so much.
People can find me online at Leo Dion.
At Leo Dion at CI am on Mastodon.
my company is bright Digit.
We have all sorts of new stuff on there.
I just posted the iOS backend cheatsheet, so if you're looking for
what kind of backend to pick, youmight want to take a look at that.
(32:03):
And yeah, it was great foreverybody to join and thank you
so much and have a great summer.
Talk to you later.
Bye.
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
The Joe Rogan Experience
The official podcast of comedian Joe Rogan.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Special Summer Offer: Exclusively on Apple Podcasts, try our Dateline Premium subscription completely free for one month! With Dateline Premium, you get every episode ad-free plus exclusive bonus content.