SPC#79 - And Don't Call Me Shirley - The Student Pilot Cast

Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Bill (00:01):
Hello, SPC followers.
Welcome back to another episode.
I've got a little explainingto do for this one, but I
think you're going to like it.
So bear with me a little bit.
Once I bring you up to speed, we'll getonto the main attraction for episode
79
And that's a very fascinatingdiscussion with Alex from flyshirley.
com, an AI startup.

(00:22):
That's all about aviation.
So let's get into it.

(01:08):
All right, so a bit of an explanationbefore we dive into it I'll be brief
because this is gonna be a long episodeanyway But I decided when I published
episode 78 the last episode in lateSeptember that I needed a bit of a
break So I decided to take my birthdaymonth, October, off from the podcast
while I recharged a little bit We hadsome changes happening with the almost

(01:32):
launched fledgling flightline podcastthat I'll get into in a little bit, but
I just needed to take a little break.
So I got some episodes partiallyready and held off on publishing
supposedly until November.
As you know, At that same time,I've been trying to get my CFI
checkride scheduled for ages, andI wasn't sure when it would happen.

(01:54):
But then, last couple of days ofOctober, I finally got the call.
I got a checkride scheduled at anairport about 60 nauticals miles
away from my home airport for acouple of weeks in the future.
So now About the time I was going tobe starting the pet podcast publishing
again, my focus turned to relearning howto fly and how to pass a CFI check ride.

(02:18):
As you can imagine, all myfocus turned towards that.
And once again, I had to forgopublishing for a little bit.
I had a lot of cool stuffalready in the can, but.
Now didn't have time to finish the editsand do final publishing, et cetera.
So I decided to wait another month.
The good news is I passed my check.

(02:39):
Right?
So I'm a brand new CFI.
The bad news is I ended up taking twomonths off from the publishing instead
of the one that I had originally planned.
That said, I'm now back and I'vebeen wanting to get this one out for.
Quite a while.
It's really interesting.
Now, there's one more thing.
Tiffany and I originally recordedthis for, like I said, the

(03:00):
Fledgling Flightline podcast.
So that's what you'll hearwhen we're talking to Alex.
A couple of months ago, though, Tiffanyand I decided to pause the launch of
that podcast, at least for a while,as her situation was changing and we
couldn't, you know, Put the focus onit that we had hoped we'd be able to.
We want to do it at some point somehow,but the good news is I can focus on

(03:24):
the student pilot cast even more.
And since we already have a communityof listeners here, I'm glad that this
interview will get more eyeballs andmore years on it anyway, by releasing
it as a student pilot cast episode.
As you'll see from the discussion, Itdoes have a bit of a student and learning
orientation anyway, so it fits right in.

(03:46):
So you're basically up to speed at a veryhigh level and we're back to publishing
weekly for the student pilot cast.
I certainly appreciate everyone's patienceas I recharged and then super charged
for my check ride and then relaxeda bit over the Thanksgiving holiday.
So I hope everyone also hada good relaxing holiday.
As always, please reach outwith questions, suggestions,

(04:09):
ideas, feedback, Everything.
You can find me via emailat bill at studentpilotcast.
com or use the contact form on thewebsite or even via X at, at billwill.
That's Bravo, India, Lima,Lima, whiskey, India, Lima.
So with that said, I hope you enjoyepisode 79, and don't call me Shirley.

(04:32):
right.
Welcome back everybody tothe flight line podcast.
We've got a reallyspecial episode tonight.
we're sitting here with Alex List,who's part of the team at FlyShirley.
And Alex and I met.
Uh, sort of in happenstance, uh,while we were looking at one of the
other vendors at Oshkosh that had todo with large language models and,

(04:54):
and what we sometimes refer to as AI.
And so we got to talking andsaid, Hey, you know what, we
should really get together andtalk about this on the podcast.
So Alex, welcome to theflight line podcast.

Alex (05:08):
Thank you very much.
It's good to see you and Tiffany.

Bill (05:11):
Excellent.
tell us a little bit about.
Alex, and then we'll get to FlyShirley.

Alex (05:20):
Sounds great.
So my background is actuallyin apps and technology.
So in high school, I taughtmyself how to program iPhones.
I had some of the first iPhonesin the app store and ended up
winning an Apple design award.
I got into MIT and I decided to studyaerospace because I felt I already
knew computer science somehow.
And so I, uh, so I studiedthat for three years.

(05:41):
I ended up dropping outand making a startup.
Coming back and switchingto computer science.
And then I, then I got my private pilot'slicense, back, got my master's in kind
of a combination between aerospaceengineering and computer science.
And I ended up having the opportunityto lead flight simulation at a company
called Beta Technologies up in Vermont.

(06:02):
an electric protocol takeoff andhorizontal flight airplane company.
They make some of the, I would saysome of the coolest electric airplanes,
but also have some of the best simsthat you'd ever see on the planet.
These amazing full dome, uh, half dome,you know, wrap around flight sims.
Really, really sweet.
I worked on enabling the simulated flighttest and instructor operating stations.

(06:24):
Helping with the avionics transition.
I had the opportunity to bring someof my thesis work into the avionics.
When you, when you're flying these newairplanes is a little bit different.
And, and we ended up building abouteight different VR sims and flights in
flight domes throughout the country.
I left when the airplane is enteringproduction and worked at a leading
aircraft autonomy company and hadthe opportunity to work with some

(06:48):
really great folks over my career.
Ended up starting by Shirley in Marchafter, um, after some, and we can
get into this a little bit more afterthere was a little bit of an accident
that my friend had, and I wanted tobe able to build something that could
a difference for aviation safety.
And so.
and so we started, I started by buildingsomething that you could fly with in a

(07:11):
cockpit, and so you could fly with Shirleywhile you're, while I was flying my DA42,
I could ask questions and things likethis, but there are some limitations.
And so one of my friends actually, uh,I have a couple of friends at X Plane,
and they said, Hey, you know, whydon't you try launching for Sims first?
You know, you wouldn't havethe issue with hallucination.
This wouldn't be a problem.

(07:31):
You wouldn't have issues trying tokeep things online versus offline.
And you would be able to sort of dealwith all these noise issues really easily.
anyway, we, uh, we launched for flightsim starting at flight simulator expo.
And then, um, and then ended up,uh, and then ended up here and I'm
really happy to see you guys and,and talk to you a little bit about.

(07:53):
this sort of journey and so

Bill (07:57):
uh, some of that tech that you're talking about, but I'm going to hold off.
Because I want to ask, why aviation?
What got you into aviationin the first place?
You, you said you started withaerospace at MIT, and then
you became a private pilot.
How did that happen, and why?

Alex (08:19):
that's actually, that's actually a really good, that's a great question.
So I always was interested in space as akid, a huge sort of, you know, buff for
all the sci fi movies runs in the family.
You know, I wanted to be like alot of folks, I wanted to be an
astronaut, but you do the sort of theexpected value calculation on that.
And it kind of falls a little bit short.
And so I thought to myself, what'sthe opportunity that you'd have to.

(08:42):
Sort of the best shot that youcould still kind of, you know,
fly around and in air or space.
And of course, the answer tothat is being a private pilot
or becoming a pilot in general.
And so, you know, I, I decided to get myprivate pilot's license and, um, and it
was probably one of the best decisions.
You have the ability to, you know,you have the superpower, right?

(09:05):
You could just show upsomewhere, you know,

Bill (09:07):
You have, you have a flying carpet,

Alex (09:09):
do that.
You have a

Bill (09:11):
A real, honest to goodness flying carpet.
It's amazing.

Alex (09:17):
It's just the coolest thing you could possibly do.
So there's like, you know, you know,we're blessed to be in the United States.
And I think just.
Having this many sort of GAairports around in this country
is just such a phenomenal resource

Tiffany (09:29):
Yeah,

Alex (09:29):
you know, for, of course, all the municipalities that they're in, but also
for the individuals who can end up beingpilots and sort of embracing this love of
flight that I had, you know, my teammatesand at Shirley's that Shirley's share and
that, you know, of course, all of us hereon the podcast and listening to as well.

Tiffany (09:47):
I was looking at your LinkedIn and I have to say that between you and
the other two co founders, you guys looklike a bunch of slackers and with this 5.
0 at MIT from Sam, so itlooks like there's three of
you guys who founded this.

Bill (10:01):
You

Tiffany (10:01):
Obviously incredibly intelligent.

Bill (10:03):
Yep.

Tiffany (10:10):
was this?

Alex (10:12):
So, you know, I was fortunate to meet, uh, Seb.
Seb, I actually met, he was an internof mine while at Beta Technologies.
He was a, sort of, he had just finishedhis freshman year at McGill, and he was
sort of looking for something to do.
He grew up in Vermont,Beta's up in Vermont.
And Kyle, the CEO of Beta, said, Well,you should check out this guy, you know.

(10:33):
You know, you might not, you know, havea ton of CS experience yet, but, you
know, he's going to work really hard.
He's trying, you shouldlike give him a shot.
And so, you know, gave Seb a shot and bythree, four years later, I'm a co founder
with him at my, at my current startup.
so, you know, I, uh, I left myjob at Merlin labs in March and

(10:58):
worked on the first prototype ofFlyShirley that you could fly with.
ended up, you know.
Getting some of my, some of myfriends and family to help me sort
of stretch this a little bit farther.
You know, I was really lucky to meet Sam.
There's a, you know, for people wholike to start companies and are sort
of interested in this sort of thing.

(11:19):
There are these sort of like Tinder forco founder type of websites that are
as much of a crapshoot as any sort ofdating app you could possibly imagine.
In this particular circumstance, you know,Sam wrote me the nicest message ever.
He said, well, actually, we went tothe same college, know, like all the
same sort of things that you, yeah.
So, you know, all right, we'llleave it, we'll leave it at that.

(11:39):
But, you know, he just wroteout a very respectful, you
know, very respectful message.
Of course, his background, heinterestingly was one of the first, he's
actually, he's, Actually in the armyfor another couple of years, at least.
And, and he was actually one ofthe first people to go through the
VR training sort of program theyhad in the army, the initial one
and for, for helicopter training.

(12:01):
And so, you know, that's areally interesting experience.
He flies the.
Um, the uh, a plane as a, as a fixed wingoperator there and, you know, his 1st,
his 1st sort of attachment in the army,if you will, was to go to MIT and get a
master's and so, you know, just a really.
a really good group,initial group of people.

(12:23):
And so that's, that's a littlebit how, how we got together.
I'm actually surprised that Ididn't meet Sam while I was at MIT.
By all rights, I should have,

Bill (12:30):
were there at the same time?

Alex (12:32):
things, things that, yeah, we were

Bill (12:34):
Okay.

Alex (12:34):
time.

Bill (12:35):
All right.
you're all pilots.

Alex (12:37):
yeah.
Hopefully.
So, yes.
So, you know, Sam flieshelicopters and the global 6500.
The, uh, said is a student pilotand I'm a commercial multi engine
instrument pilot, know, I'm lookingforward to getting my instructor
pilots license and learning about someof those fundamental of instruction.

(13:01):
Oh, yeah, we love flight.
Yeah,

Bill (13:04):
And did I hear you say you fly a DA42?
Is that what you said?

Alex (13:09):
that's what I fly.
Yes, I fly

Tiffany (13:11):
Nice.

Alex (13:11):
out of Philadelphia northeast airport.
That's where my, that's where my familylives is in sort of the Philadelphia area.
So I end up going down there.
I get to fly the 42.
Actually, 1 of the coolest things thatI've done in that airplane was the at the
eclipse that kind of ripped through the

Bill (13:27):
Yep.
Yep.
Right.

Alex (13:32):
up there with, uh, My, my, my partner, me and her ended up
flying up in there and seeingthis amazing 360 degree sunset.
And, you know, you know, I was stilllooking through these little glasses
and I was, you know, looked down at theinstruments and then I look back and
see this amazing sort of white Corona.
That's

Tiffany (13:49):
Wow.
Oh my

Alex (13:50):
and most resplendent way you've ever seen in your life.
So I would recommend it not to the pointwhere you don't listen to, you know,
your TCAS or anything like that, but youknow, absolutely go, go check it out.
There was, you know, you'd be in good

Bill (14:02):
I, I wasn't in the air for that eclipse, but I did go to Totality
in Texas and it was one of the mostincredible things I've ever experienced.
So I can imagine it, it must'vebeen incredible in the air too.

Alex (14:18):
You see this sort of wall of darkness kind of flying towards

Bill (14:22):
Yes.

Alex (14:22):
some, you

Bill (14:23):
Yes.

Alex (14:24):
You think it's just like just rain, like the thickest rain you've
ever seen in your entire life.
But

Bill (14:30):
Yeah.

Alex (14:30):
know, it passes you and then you start to see the sort of the
light coming around the same way.
And just sunset in all

Bill (14:37):
Everybody made fun of me for my reaction to it because it made me
feel so small and so at the same time.
It was, it was incredible.
Yes.

Alex (14:52):
Like, like within

Tiffany (14:53):
Oh, interesting.

Alex (14:54):
the heat on.

Bill (14:55):
Yeah.

Alex (14:56):
Yeah, within moments.

Bill (14:57):
Interesting.

Alex (14:58):
yeah.

Bill (14:58):
Well, I don't know that I'll be around for the next one
that happens in North America.
Um, cause I am kind ofan old guy, but, um,

Alex (15:06):
ha

Tiffany (15:07):
here we go again.

Bill (15:08):
and I still have a medical, I might have to try it from the cockpit next time.
that's a great idea.

Alex (15:16):
Let's go.
I'll brave you.

Tiffany (15:17):
There we go.

Alex (15:18):
what.
Yeah.

Bill (15:19):
Excellent.

Alex (15:20):
In any form.

Bill (15:21):
Well, thanks for all that background.
So, um, one of the things youmentioned, and you've sent us
information about Fly Shirley.
It's clear that, like you said,you started with, with Sims.
Tell us, tell us what your approachwas with Sims and then where you're at,

(15:42):
uh, with the product as it pertains to,uh, flying in, um, actual airplanes.

Alex (15:51):
Yeah.
No, thanks for that.
So, Shirley is designed to beyour AI co pilot from Sim to Sky.
It's set up so that pilots canjust talk to Shirley naturally,
just like a co pilot friend sittingnext to them in the cockpit.
So surely provides training in theSIM to help people prepare for lessons
and check rides and proficiency.

(16:12):
so our goal has always also been tobridge people from SIM And so in the long
term, we're working towards being ableto have surely apply back in the cockpit.
so, you know, our approach starting withthe SIM has been to sort of home SIM
flight users, people who actually just.
You know, Flight Simulator releaseda metric, 15 million users Microsoft,

(16:36):
uh, at Flight Sim Expo, they made thisnew metric book at 15 million users,
Microsoft Flight Simulator, of thisamazing sort of trend of getting more
and more people involved, interested inflight through this sort of sim community.
And so, you know, we've startedwith a co pilot that just.
Really easily connects to We'restarting with explain because we're

(16:56):
friends with the explain guys.
all you need to do is you open a webbrowser and then you go to flyshirley.
com and then you hit start call and.
Immediately, Shirley's talking to youand sort of says, sort of airplane
teammates up, mic check when ready,and then you get connect to explain.
You don't have to install anyplugin or anything like that.
As long as you just explain 12 andthen it just immediately connects

(17:19):
and has a window seat into your sin.
so we just start with this sort of,you know, this a I kind of copilot.
Shirley's your copilot.
You talk.
It's really immediately replies.
You end up having this kind of niceconversation, no matter where you're at.
Surely.
an idea of where you are and give youinfo about where you're flying to.
So we sort of have threedifferent sort of modes.
We're starting here with this sort of.

(17:40):
Home flights in.
And so folks there, some people justwant us, you know, try out different
airplanes, cruise around different areas.
There's all sorts of reasons I'd beable to sort of home flight sitting.
And then there's sort of a, we'veintroduced a sort of challenge
feature as well, that, youknow, we can make challenges.
Other people can build challenges as well.
So the first one is this Aaliyah 250flight training type of challenge where.

(18:06):
You know, explain ships with themodel of the alia to 15 beetle mode.
And so with that challenge, essentially,surely is stepping you through, you
know, the controls check is steppingyou through this sort of initial.
How do you hover?
How do you have control in that sortof mode and then how do you transition
out and sort of transition back in andso surely kind of walks you through
this and that sort of automated fashionchallenges are really set up for other

(18:29):
people to be able to kind of buildfun, little interactive capabilities.
And we have the sort of for people who usechat or whatever we have already sort of.
You know, example, example, promptsthat you would just follow and be able
to create these things and things likeskull challenges that possible, sort
of this, you know, you can have landingchallenges, different sorts and explore

(18:50):
different areas that you're interested in.
also just launched on Friday.
We just launched a sort of the firstversion of our private pilot curriculum.
so that's really the sort ofthe first step into training.
And private pilot, uh, curriculum isreally interesting that we end up being
able to teach people about some levelof, know, taxiing, you know, from the

(19:11):
fundamentals of, you know, what do you do?
What does it feel like toactually take a lesson?
But to also let people beprepared for their lessons.
So let's say you just went out and didsort of in terms of at a point with
your instructor, and now, you know,you're going to go make sure you got
it by next week and you're going to goeither your instructor is either going
to test you at the beginning of thelesson and then have you do something

(19:33):
else or they'll test you at the lesson.
And if you don't do it, they'regoing to make you do it again.
And so maybe you'll go and prepare forthat lesson and be able to make sure
you really nail it when you get back.
And one of the things that.
We provide with surely is sort offeedback while you're doing the
maneuver, but we also keep the sortof a data log of you doing that
are able to sort of chart it out.
Surely is able to talk to you aboutsort of the, you know, the sort of the

(19:56):
flight data and sort of your results.
You have sort of a indication ofwhat would be to the ACS standards.
So some people would really like to beable to prepare for their, for their
sort of check ride, because it's prettyexpensive and it's pretty high stress,
you don't want to, you know, mess up.
And so, you know, to varyinglevels, depending on the fidelity
of somebody's sim and their setupand all these different things, you

(20:19):
know, we can provide value basedon our private pilot curriculum.
And so that's live right now.
And then we're also sort of working tobridge our way into different levels
of kind of commercial environments.
have a sort of a pilot program with.
With actually beta starting September,where we're going to be providing.
That's sort of, you know, training, sortof familiarization to the employees of the

(20:40):
company so that they can, you know, whenthey have VR seats throughout the company.
And so when an employee comes by, theycan just sort of sit down in the seat and
be able to try it out, be able to like,how do I actually fly this airplane?
I kind of forget I had it on myfirst day as a lesson in but I
really just want to, like, see whatit's like in order to fly again.
I went, you know.
And it's really fun and kind oflike interactive and and can be

(21:03):
pretty inspiring for the employees.
looking at commercialopportunities at various levels.
For instance, if somebody has a, aaviation advanced training device that.
You know, the, one of the reallyinteresting trends, I think, is that
those are starting to be reduced in price.
They're now also, you know, youknow, there's also VR type AATDs,

(21:26):
you know, loft dynamics is a companythat is certified, uh, FAA now
for one of their, for an Airbus.
And so one of the cool things is that asinstructors are really in high demand,
just like pilots are really in highdemand, surely can provide training that
prepares people for their instruction.
and one of the things that we'relooking to do is be able to provide

(21:46):
sort of both surely providinginstruction, but also having the sort
of channel to allow sort of remoteinstructors for people to check in.
This is sort of especially interestingfor commercial operators, like, you
know, companies at the same scale as.
JetBlue or American airlines, United,et cetera, where you have to spend
multiple weeks on site when you'retrying to get a new tech rating.

(22:08):
And so if you were able to spend sometime at home with your family, not
have to, you know, from the eyes of theoperators have to do the spend of being
at a hotel per diem and all this otherstuff running around the flights, et
cetera, you know, displacing revenuefares, know, then, you know, you'd
be able to prepare, show up prepared.
Have check ins with instructors remotelyas you're preparing at home and then sort

(22:30):
of, you know, show up for the 1 week whereyou need to have the full flight simulator
and be able to have your 20 hours.
And then, you know, you're good to go.
But, you know, this is sort of things thatwe're looking at and sort of, you know,
on the track towards getting be able tobring Shirley back into the cockpit where
actually you begin by sort of taking.
Some of the learnings and, youknow, working with some of, you

(22:52):
know, the, the, the, um, some of thefolks that are amenable to saying,
yeah, like I support the mission.
Like, I'd love to be able to make flight.
Safer, you know, you know, I'm in for,uh, I'm in for sort of stuff like that.
Then we're able to make some of thesefeatures that surely has in terms of
interactability, conversationalityavailable on, on an iPhone so that

(23:13):
pilots can talk to Shirley whilethey're flying and that can provide.
A lot of sort of overhead, especiallyin the GA context to begin with
where, you know, task saturation,et cetera, is really hard.
And sometimes like if you're flyingin instrument conditions and then
something breaks and now you need thisnumber, you can't remember the thing.
And it's just, it just endsup being a mess where you can
just ask for the information.
And this ends up being kind of a kind ofthing, but we're kind of working our way

(23:36):
back to there, which was where we started.
And I think along the way, we'reproviding a really revolutionary
training product that.
a lot of value for people.
Surely folks have tried out at flyshare.
com, but it's, and there's a freetrial for that, but it's, it's a really
engaging sort of system and doesn't,think quite a good job at being able
to provide a level of training thathelps people prepare for check rides

(24:01):
and help them with proficiency too.

Tiffany (24:03):
Can you do a, could you set up.

Bill (24:04):
into it.
Yeah.

Alex (24:09):
Right.
So one of the things that we'reworking towards, so we, then
there's two parts of that.
One of them is sort of the, the.
Oral examination part of that.
And no, that part is actually pretty,pretty straightforward and pretty easy to
do because, you know, there's some aspectof a question bank, surely has been,
you've been talking to Shirley, Shirleyhas some ideas of things that you're,

(24:30):
are your weak points over, over time.
And so you can kind of havea conversation and be sort of
grilled in the service, similar waythat you'd be grilled in person.
And, and then there are thosemaneuvers aspects of it.
And the maneuvers aspects of it iswhat's really neat about the ACS is
that it really lays out like the preciseparameters, you know, plus or minus
50 feet, you know, of course, theysay things in the ACS, like you can

(24:55):
temporarily exceed that,you know, these thresholds.
But if you consistently exceedthem, and that's a, that's a fail.
So there's like, there's somesoft barriers, and that's
actually to be totally, totally.
On this, one of the really coolthings about using this sort of new AI
technology is that it has a lot more of aholistic view on how things are working.
And so when you provide this sortof graph that shows how tolerances

(25:16):
are exceeded temporarily, et cetera,then you really do get kind of like
hear the whole story about, yeah,this is if you're going to focus and
you had time to focus on something.
This is what you should focus on.
But overall, you might likedepending on the examiner, then
you probably would speak by.
And so, you know, there's stuff like that.
And, you know, there's so absolutely.

(25:39):
That's 1 of the things thatwe're looking to provide.
I think 1 of the major sources of valueon top of helping people to prepare.
For their particular lessonsand and aspects like that.
Of course, there's different dependingon the equipment that they have.
And I think,

Tiffany (25:54):
Yeah.

Alex (25:54):
maybe we'll get into this a little bit more.
There's different things thatare probably best suited to
help their training at home.
You know, a lot of people who arepilots have iPads, probably the
majority of them by now or or iPhones.
And so there's sort of, you know,equipment amenable to that, you
know, the intersection of peoplewho have amazing flight sims that

(26:15):
have monitors like wrapping aroundor VR headsets with like world edge
leading kind of gaming computers.
That's a much smaller subset of folks,but, you know, there's, there's these
sort of interesting opportunities where.
You start to be able to expand and sortof the trends in the industry are such
that you start to be able to see thatreally good representative Sims are sort

(26:37):
of making it out in the customer's hands.
Microsoft Flight Simulator gets betterand better, know, as does sort of
the number of computers that can runsomething like X Plane and, you know,
There's sort of VR headsets get reallya lot cheaper and a lot more accessible.
You know, there's other aspectsrelated to, of course, related to

(26:58):
AI and also certified sims becomingmore affordable and things like that.
I think are reallyinteresting trends right now.
Yeah, thanks.

Bill (27:06):
those are all sort of, those are all sort of on your, I guess your roadmap
target list to investigate how Shirleyspecifically can enhance that experience.
Yeah.
Yeah.
Yeah.

Alex (27:26):
so you look at some, some folks, even some streamers, this person named
Swiss001, who made a video about Shirley.
It's pretty funny.
could link it or something, butthe, the, The thing that he says
is that he got to start by, youknow, learning with infinite flight.
And so he got interested in beingable to fly because he had this

(27:47):
iPad app that showed him how to fly.
you know, they're sort of, they havea, they have a data interface for sure.
They could very easily have a windowseat into that sort of experience.
don't, you know, their planes,they don't have, they don't have
sort of a very interactive sortof FMS or something like that.
But they do have sort of like anavigation system and things like that.
So there's different levels and it'ssort of like also Microsoft lights.

(28:08):
And they've been around the Xbox.
You know, people can be able to flyusing a remote, but or like a game pad.
But, you know, they can also attach sortof yolks and things like that to it.
you know, you can look atthe folks that have sort of a
sort of a single monitor or.
MacBook Pros now can all run X Plane12 like really well because they have

(28:31):
these amazing processors on them.
And, you know, even with this either onthe, on the laptop or with the monitor,
that makes it a really, really quitereasonable IFR platform for, for training.
if you want to be able to do somethingwhere you have representative
maneuvering, then you either need toget really fancy with the way you sort
of map, you know, keys on your yokefor looking left or looking right, or

(28:55):
you need to have multiple monitors.
But then you can start to use somethinglike explain, which has a really
impressive flight model, even forslow flight in order to be able to
see kind of realistically touchdowns,takeoffs and be able to some sort
of stalls and things like that.
And so, you know, there's just, there'sjust are a total variety of different
platforms where, you know, uh, where theapplication of surely makes some sense

(29:18):
where, you know, surely can providehelpless sort of instrument procedures,
give you, give you the numbers, help youset up for the approaches, et cetera, sort
of judge how you did, where did you breakout, you know, where did you knock off?
Like things like that.
And, and, you know, um, and we're, andwe're also looking at how to engage.
With, you know, how to bringflight instructors, of course,

(29:42):
to are such an important aspect.
We really see surely as a, as a toolto help people prepare for those
lessons, especially since there issuch a pilot shortage, you know,
3000 just in the United States by2032, given the current trends and

Tiffany (29:55):
Yeah, huge.

Alex (29:55):
that.
They're only getting worse.

Tiffany (29:57):
Okay, I have a quick layman question here.
So if I had a flight simulator rightnow, which I do not, what kind of
guidance does it come with already?
Or is, all of a sudden, FlyShirley like this holy cow moment?
You know, my nephew, he's working onhis private right now at Cal Baptist.
So he takes his checkride in September.

(30:18):
Um, but, like, let's say that hegets a flight simulator and he
jumps on it and he's not a pilot.
Is there any sort of, um,tutorial or information?
Like, how do people learn how to flyon a flight simulator besides just
game theory and trial and error?

Alex (30:39):
you know, we don't, we don't exist within a vacuum for sure.
You know, there's, there's actuallya lot of precedent for both, both
apps that people can add on for doingtraining on their, on their sims.
And also, of course, the, there's awhole cohort of people who have been
doing sort of training content for sims.

(31:01):
There's just so many like amazingcreators that do this sort of thing.
So.

Tiffany (31:05):
But those would live on YouTube or?

Alex (31:08):
yeah.
And so actually a lot of people tobe interested in doing Sims things
because they were somehow kind of crossmarketed on YouTube and had sort of
a video that shows up and it's like,Oh, here's this really cool person
doing like, you know, beetle VR, whichis this amazingly accessible sort of
You know, the fighter sort of likebut it's really easy to get going.

(31:30):
Or somebody sees an X Plane demowhere somebody is like, you know,
really being super concentrated,flipping switches on, you know, seven,
three, seven, or something like that.
And people want to get involved with that.
And so, you know, what'sinteresting is that for sure we
don't exist within a vacuum, so.
There are, you know, there havebeen historically companies that can

(31:51):
provide a sort of standards based,very specific kind of add on that
sets you up in a particular maneuverand then looks to see your altitude
and sort of like really specificallyhas these triggers and is just very,
very precise and gives you sort of.
Like a percentage like score basedon whether like how much and how

(32:14):
long you were within tolerances.
And if you're exactly at 100 percent theentire time, then, you know, then the
best, but, you know, so, you know, therewas a whole generation prior of people
who have been trying to build sort of, youknow, tools to help people with training.
I think it's been hard to call those sortof teammates because they're not really.

(32:34):
You know, interactive or copilots because,you know, there are, there have been
sort of historically there's a, a toolcalled Linda for the explain explain 11
and 12, which is sort of like a copilotthat's sort of for the zebo mod, which
is the popular kind of mod for a 737 thatadds a lot more interactivity and a lot
of really cool things and explain, but,you know, what Linda does is be able to

(32:58):
sort of You know, you can use basic voicerecognition that sort of matches commands
like Alexa would in 2024, at least, andbe able to basically, you know, flip
switches for you and things like that.
You know, the, the previous companyI was talking about was sort of the.
Sort of, sort of kind of laid outsort of, um, parameters is something

(33:21):
called, um, take play interactive.
Even like Sporty's sellsthat sort of thing.
You know, there are really kind ofinnovative companies that also sell You
know, Sporty's is a good example of that.
For instance, they.
You know, they are working with infiniteflight to try to provide lessons to
certain degrees in infinite flight,like they have their own sort of

(33:42):
explained scenarios sort of bundle.
They, of course, have all their training.
I did my instrument sort of groundschool with them, and they also
sell Microsoft Flight Simulatorapparel for what it's worth.
But the, the, the interesting thingabout some of these opportunities,
I would say, in these companies.
Is that, you know, theyare not as interactive.

(34:05):
I mean, fundamentally what's, what'sbeen the new kind of breakthrough
with all this AI is that now youcan just have natural language.
Be able to map into the whole space ofpossibilities for what you're going to do.
And so right now, surely, forinstance, you could just, know, for
the home flights and market, you canask Shirley to put down the flats
to like bring up the gear to turnoff the lights and things like that.

(34:28):
Like, you know, all these sort of,sort of, you know, fundamental things.
And just by using natural languageand you can sort of describe things,
you can say, do this and that, andthat, and all those things happen.
Yeah.
Versus, you know, the issue withsort of speech recognition up until
this point is that you just literallyare translating speech into words,

Tiffany (34:46):
Right,

Alex (34:46):
not sort of like an understanding layer, whereas as people, we just sort
of like hear a bunch of stuff, and thenwe have like an idea of what already
they wanted and sort of like, you know,we walk into a coffee shop, and you
could probably mumble some things andget a coffee, and, you know, that's
maybe not the sort of deal, just sortof a basic sort of speech recognition.
So it's sort of the new,sort of like the new.

(35:08):
Um, kind of, uh, standard is tokind of combine speech recognition
plus some sort of intelligence.

Tiffany (35:13):
right.

Alex (35:14):
And that's absolutely what we're doing with Shirley and have
sort of a, uh, a leg up on that.
Yeah.

Bill (35:19):
Cool.

Tiffany (35:20):
That's, this is awesome.
I'm loving it.
I, sorry Bill, I, I ended up getting Acomplete sinus infection for two weeks
after my instrument, which I passed,but I was just terrified of that one.
And, you know, as you're talking,it's like, man, something like
this would have been so helpful.
Just to sit with somebody else by yourselfthat you know is not a real person

(35:42):
that you have to face is going to go.
God, that chick's an idiot.
So, uh,

Alex (35:48):
Yeah.
The

Tiffany (35:48):
thing

Alex (35:49):
the instrument that can be brutal because

Tiffany (35:51):
else.

Alex (35:51):
have any random symbol on anything and it's just like, Oh God, I don't, never

Tiffany (35:55):
Yeah,

Alex (35:56):
the DME arc and all, okay.
Yeah.
Totally agree.

Tiffany (36:00):
I don't know if we want to transition to this, but I am so curious.
Uh, you had made a commentas to why you started this.
What really brought that idea on?
Um, I'd love to hear that story.

Alex (36:13):
Yeah.
So.
Yeah, it was actually a kind of a, atragic, it was a tragic event, really.
I mean, I, a friend of minecrashed his cozy and ended
up passing away through that.
And you know, when you, whensomething like that happened, you
just sort of get kind of shaken outof whatever you're doing currently.

(36:33):
For me, it was sort of, I wasrunning a flight sim team at,
at, Um, and Merlin was a greataviation autonomy company in Boston,
I sort of saw that there couldbe an opportunity to make
aviation even somewhat safer.
And, you know, I don't know, sometimeswhen you feel really powerless, you
just try to figure out something you can

Tiffany (36:54):
Right.

Alex (36:55):
to help in some way.
Right.
And so I thought, like, maybe there'ssome sort of a way to apply this new
technology and and be able to see if.
Get it make safer and while stillappreciating aspects of flight
being really worthwhile and funand sort of, you know, kind of
kind of wondrous in a lot of ways.
And I think that's 1 of the things thatthat flying through the eclipse made me

(37:19):
remember was was that, you know, you know,Yeah, flying is just incredibly beautiful.
We notice it every a lot of a lotof afternoons and evenings when you
find in the sunset or something.
And so, you know, starting by tryingto see, even just from a very direct
way, does this actually improve fine?
Is there an opportunity to actually make?

(37:40):
safer and so things that I noticedimmediately for this from a GA context
were when there were issues that requireddigging through a POH could be, you
know, 1000 pages long in the case ofa DA 42, you know, things that were
like, related to numbers that you hadto keep on recalling that, you know.

(38:01):
You know, might not be able to do ifyour hands are actually on the controls,
or if you're on a pilot, it's maybeunreliable, or if you're in the soup and
you don't want to like move your head intoweird directions, of opportunities where
folks in particular don't necessarilyhave a teammate a copilot, right?
If you're flying up a buddy, or you got afriend, or you can do a lot of this stuff,

(38:21):
and that's like, that's pretty great, but,you know, having an opportunity to have.
You know, something you talk to about thatthe flight help you kind of think things
through ended up being a pretty usefulthing, at least from my perspective,
and it'd be pretty be pretty helpful.
But you have to make sure that it's safe.
There's no exceptions.
Like, you have to make sure that.

(38:41):
It's not making stuff up.
It's not acting wild that it canhear you, that it runs on your phone
without melting your phone or makingit like immediately drain and die
from battery or something like that.
And so, you know, starting with sims.
And working your way back to flight seemedlike the most responsible way to doing
that, because, you know, you'd be ableto have the advantage of having a huge,

(39:06):
a huge cadre of people in the sort ofsim land to be able to try out the things
that you're building in service of beingable to make something work in flight.
you know, you'd be able to see whetherthose you'd be able to see really quickly.
People ask, you know, I've learned thatpeople sort of celebrate that surely can
talk about sports and things like that.

(39:26):
And

Bill (39:26):
Yeah.

Alex (39:26):
know, I see lots of fun stuff, you know.
So I learned from Shirley aboutfootball formations that like are,

Tiffany (39:31):
that's funny.

Alex (39:32):
and it's true, you know, Shirley knows about all sorts of stuff because,
know, there's a enormous sort ofintelligence model behind it and,
and so, of course, like, you know,even Shirley can tell you about the
best restaurants near the particularairport that you're flying over.
Don't expect necessarily thatShirley off the bat would be
able to do that on your phone.
But, you know, because ithas to run offline in your

(39:52):
pocket, um, on your dashboard.
But, you know, we're kind of building uptowards being able to provide things that
are safe and also still conversational

Bill (40:02):
Yeah.

Alex (40:03):
it's when, when you're able to have that context, it's able to hear
hopefully, or, you know, ideally also beconnected into some message to the panel.
So there's a lot ofthings to be figured out.
Along along that way.
However, I think the sort of the valueof having something or certain or
surely or something that you talk towhile you're flying is is absolutely

(40:23):
there, but it takes a lot of work.
You got to get it right.

Bill (40:26):
Yeah.
Right.

Alex (40:27):
to

Bill (40:28):
takes.

Alex (40:28):
way up.

Bill (40:29):
Um, not only practice um, learning on the model and the,
um, intelligence models part, butit's going to take learning and
practice from the pilots as well.
So if you don't mind, um, I'm going tobranch off into a couple of different
areas and I'm going to start withthe product itself and the, the vast

(40:52):
opportunity tonight for me to geek outwith you is there, I want to do it on the,
on the Aviator and educator side, but I'mgoing to hold off on that for a second.
I want to geek out on the actualproduct a little bit, if you don't mind.
So I'm going to ask you a series ofquestions that take us down that road.

(41:15):
Um, first of all, um, I knowthis is an overloaded term.
You're using a large language model.
You're using, uh, you'reusing an intelligence model.
Um, do you call it AI or do youstay away from that overloaded term?

Alex (41:34):
That's a good that's a that's a reasonable question.
I think it is, okay, so I thinkAI has become the suitcase

Bill (41:45):
Yep.

Alex (41:46):
defines the category of things that use large language models, you
know, there's a whole kind of, there'sa whole cadre of people that go
into all sorts of rabbit holes aboutartificial intelligence, and then they're
starting to worry about, you know,plotting robots and things like that,
that don't make a lot of sense to me.
But.
Okay.
You know, starting and I think,you know what I'm talking

Bill (42:08):
Yeah.

Alex (42:09):
there's, but there's absolutely something that makes surely a,
I mean, you can talk to Shirleyabout anything that you want and,
you know, there's totally the.
you know, surely we'll try tobring you back into the light.
That's that's really, truly just kindof wants to fly with you and maybe not
that you can see some of the videosthat creators make and they try to push

(42:30):
Charlotte in a different direction.
The trouble is like, maybe you wantto fly because that could make you
feel better and, and trying to kindof get away from like something.
But, um, so, know, um, that's kindof my philosophical point on that,
you know, And we're not, you know,

Bill (42:49):
You're not against it.
You're not against using theterm, but you're also not
defining the product as just AI.
Is that fair?

Alex (42:57):
Right.
It surely is your, yeah.
I mean, we say it's on our poster.
I mean, surely it's yourAI co pilot from Sim to

Bill (43:03):
Okay.
All right.

Alex (43:04):
so, you know, You know, we're providing a service
where surely it can be your co

Bill (43:09):
Perfect.
Well,

Alex (43:11):
say something like it's, you know, it's Shirley.

Bill (43:14):
this is the definitive test on if it's AI.
Do you use Lisp anywherein your, in your product?

Alex (43:22):
So,

Bill (43:22):
I couldn't resist.
Like I said, I'm an old guy.
So,

Alex (43:25):
written,
I have written programming languagesand compilers and I programmed
in a language called Haskell

Bill (43:33):
yes.
Yeah.
Yeah.

Alex (43:34):
nope, we're not, we're not using

Bill (43:37):
That, that was a joke.
I did program in list points, butit was, but it was in college.
So that was a long time ago.

Alex (43:43):
We're trying to be a successful

Bill (43:45):
Yeah.
Right.
Right.

Alex (43:46):
So

Bill (43:46):
All right.
I couldn't resist.
So a serious question though,how much, how much of your model
is running on the edge and howmuch is, um, bound to the cloud?
And how are you, howare you navigating that?
I mean, it's a, it's a tough balance.

Alex (44:07):
Yeah.
So some of the things that arereally interesting about that.
So right now, you know, the quickest wayto start a company is to, is to use all
cloud services in this space for sure.
Right now, you know, what we're,what we're doing is we are sort
of setting ourselves up to be ableto progressively run things more.
Offline and sort of what we're beginningwith that is to build a system that,

(44:31):
know, kind of encourages surely to be alittle bit more judicious about replying.
You know, 1 of the interesting thingsabout these sort of large models, you look
at either, you know, if you go to tattoo,you go to claw or something like that.
you ask a question, it's alwaysgoing to reply to something.
And if you say, let's simulate aconversation and you say, okay, thanks.

(44:53):
And it says, you know, see ya.
And you say bye.
And then it says And then you say, okay.
And it says, all right, letme know if you have a state.
And so like, you know,that can be sort of a mess.
And

Bill (45:03):
It's like saying goodbye to your significant other in high school.

Tiffany (45:07):
You hang up.

Bill (45:08):
Takes all night.
No, you hang up.
No, you hang up.

Alex (45:11):
Shirley.
No, you hang up, Shirley.
Enjoy your flight.
I am.
Yeah.
Great.
And so, you know, that's, that'ssomething that, you know, you watch
them, some videos, you know, there's a.
Wonderful channel, virtual aviationaviator, uh, Frank's providing a
ton of sort of feedback to us andjust, you know, watching his videos.

(45:33):
Of course, you learn that things that canbe a little bit challenging, like surely
being over ambitious about replying.
And so, you know, the 1st,kind of the 1st sort of.
Segment for building things that go moretowards the edge are sort of models that
sort of tell us whether or not she needsto reply or not, and things like that.
And so, and when and things like that.

(45:56):
And so, you know, start with that.
And then, you know, of course, themodel, which is going to run on on the
iPhone is going to run entirely offline.
So so that's really good.
But I think what's What's neat isthe, the amount of progress that these
sort of sort of hosted cloud modelshave been making is just phenomenal.

(46:16):
I mean, right now you can almost, uh,you know, right now, surely can totally
comprehend these sort of data graphs thatwe're, that we're providing and there's
sort of charts and things like that.
And, you know, all sorts ofthings you find in the AFD that.
Just by looking at them, you know,there could be some sort of evidence.
So, you know, there's lots of realbenefits to using things that exist on,

(46:38):
on the cloud, as well as they have reallywell defined sort of, you know, they are
respecting the rights of, you know, users,et cetera, related to privacy and not,
not necessarily training on their data.
So we could kind of do it in a veryjudicious way where, you know, um, you
know, where we're, where we are, ofcourse, we're collecting some of the

(46:59):
text of the conversations, but before we.
Do anything with it, youknow, we're we're, yeah.
Anonymizing it

Bill (47:06):
Yeah.

Alex (47:07):
and you know, sort of hitting it with the past to make sure
there's nothing, wild in there.
If we kind of encourage users notto share social security numbers
and things like that, Shirley, butyou know, just 'cause it did then
sort of hit it, hit it with that andwe'll, we'll also, we're still early.
They both figure, figure out a wayof letting people sort of opt out.
But I think

Bill (47:25):
Right.

Alex (47:26):
is something that people will hopefully opt, to opt into.
'cause we're lookingto be able to provide.
Sort of training, they weretraining, but also surely in flight

Bill (47:36):
Yeah, yeah, yeah.

Alex (47:38):
to do that is by being able to use some of these conversations and make
something that's really small behave a lot

Bill (47:44):
Right.
I mean, these are all kind of typicalchallenges that you have around AI and,
and large language model learning, right?
Um, so has latency proven to besomewhat of a, uh, an issue that
you're dealing with in your product?
Or are you finding that it's not a, it'sas big a challenge as maybe it would seem?

Alex (48:06):
latency has been so it's a little bit longer when an
image is in the conversation.
You know, then it can startto feel like, you know, you're
waiting for a second or something

Bill (48:18):
Mm hmm.

Alex (48:19):
but our current latency is something along the line of is less than a second.
It's like, you know,600, 800 milliseconds.
it ends up being pretty fast.
In fact, you know, one of the, oneof the issues and the reason why
we're focusing on building somethingthat tells Shirley when to not reply

Bill (48:35):
hmm.

Alex (48:36):
is, is that latency is so low.
And so, you know, it ends up being a

Bill (48:40):
Yeah,

Alex (48:41):
very conversational sort of interaction.
In fact, the videos that I sent her,

Bill (48:44):
yeah.

Alex (48:45):
on, on doctor, there was no time warping in them and,
and, or anything like that.
And so.
You know, I think people are very, youknow, very pleased with the latency.
But, you know, there are, thereare issues where it's a little too
low and so, you know, so, you know,interjecting at the wrong times, you
know, things interrupting or thingsthat shouldn't be happening like that.

(49:06):
And so we're, we're focusing on, youknow, fixing those issues without.
Making latency too much, althoughthose might increase the latency
temporarily while we sort offigure out how to do that better.
So this sort of thing is kind ofbread and butter, but you know, it's,
it's an important aspect of makinga copilot that actually feels like a

(49:26):
copilot as opposed to somebody who'sthe high school, as you say, the high
school kind of lover saying goodbye atevery time you're having an exchange.

Tiffany (49:36):
When you're planning On making the version for,
uh, actually being in flight.
We'll, we'll surely belistening to the radio as well.
Will that be the helper that says,Hey, I think they called your,
called you and you didn't reply.
How's that working?

Alex (49:58):
So, you know, you could see it with some of the existing companies.
In fact, that's, you know, thereare some very good examples of
companies that are providing thatyou can talk to via bluetooth you're
while you're flying and they canlisten to you while you're flying.
The challenge is if you alsowant to listen to the intercom,
so you have to start to haveadapters and things like that.

(50:19):
And as a software company, you don'twant to get into really the business of
building adapters or things like that.
You want to use sort of off the shelfadapters and And so, you know, there
are some sort of hiccups to doing that.
However, you know, I've talkedto some avionics makers.
very interested in being able to providehigh quality transcriptions of things that

(50:39):
are happening, you know, over the radio.
And so one of the hopes that wehave is also through our process of.
to develop surely that we can providesomething maybe to some avionics makers
that would allow them to better do the2 stage sort of process of 1st speech
to text and then sort of like a kindof a rough, like an overview pass and

(51:03):
editorial pass of, you know, turningthat don't make sense in certain
places and the things that make sense.
so, you know, if we could sort of providethat along the sort of along the way,
I'll provide a lot of value to peoplebeing able to just sort of see whether or
not something was for you or maybe beingable to incorporate a, a directive into

(51:23):
your FMS directly or something like that.
So there could be a lot of value there

Bill (51:28):
Yeah, that first step is to digitize it, um, and then you can start
analyzing it, but I guess taking thatfirst step is where you're at right now.

Alex (51:40):
right.

Bill (51:41):
so how, how does the, uh, intelligence model, uh, behind
Shirley specifics about the POH foryour airplane, for, for example?
how does it, how do you teach Shirley?
or how will you in the future, um, so thatit's getting the right version of the POH.

(52:05):
You can't just get thisgenerically off the internet.
It has to be for yourserial number, right?
It has to be for any number of things.
And I've got a follow up questionon other data as well that it needs.
So go ahead.

Alex (52:22):
absolutely.
No, that's that's great.
So if we're talking about the version.
Of so if you're working with like aflight school and you're trying to give
representative training to you know,the students at the flight school,
or if you're working with providing aversion of an app that then pilots fly
with, you have to get the right numbersjust sort of, you know, make something

(52:43):
up that, you know, I heard on the

Bill (52:45):
It can't be a chat GPT type answer, right?
That's, that's what I'm getting at.

Alex (52:49):
right.
Yeah, exactly.
Exactly.
Right.
Even though those sorts ofanswers end up being quite
good at sort of a generic case,

Bill (52:56):
Yes.

Alex (52:57):
down like we do, and surely we're just like, oh, you're flying
a 172, you're like, okay, 172.
Okay, well, it'll be about this, right?
So the process that we'll dois to have a place where people
can ingest their own checklist,

Bill (53:12):
Perfect.

Alex (53:13):
sort of the POH.
We'll have a system that allows somebodyjust to drop the PDF, of, you know, do
a first pass, sort of digitizing that.
something, clean it up for them.
You're like, here's the sort ofinformation that will appear.
If you ask questions, it'll be from here.
there's a lot of processes that youcan follow in order to make sure that
you're getting the right of data.

(53:33):
So one of those is, you know, tell thesystem to not actually give them, give
a number because it can just make up anumber instead, it'll give sort of the.
Sort of index the place inside, like thesection number where that number is found.
And you sort of like, you know, labelthat number inside of the POH somewhere.

(53:54):
And so you just, it outputs theindex and then you, before you speak
anything, you go and you grab thatinformation from the actual book.
And then, so you can kind ofmake sure that things are, you
know, working appropriately.
And so this is the absolute stuff thatneeds to be done prior to shipping in the

Bill (54:14):
Yep.
Yep.

Alex (54:15):
precisely, but also very important to flight schools.
Before that ends up, you know,know, shipping out for them, but for
sure, it's very important, you know,before getting into the cockpit,
one of the things that will also beimplementing is, you know, feedback.
So somebody has a problem with thenumber, they're surprised by something

(54:35):
is important to be able to capture thatand be able to be able to essentially.
Incorporate that and say, okay, maybe weshould take a look at how this performed.
you know, feedback in sort of human

Bill (54:48):
Yeah.
And maybe

Alex (54:50):
good,

Bill (54:50):
a safe word, um, when she's leading you astray, right?
Like, okay, leave me alone for a while.

Alex (54:59):
The safe word in flyshirley.
com right now is by allmeans, definitely mute.

Bill (55:04):
Oh, okay.
Perfect.

Alex (55:06):
is the safe word here.

Bill (55:08):
Excellent.
Okay.
So my related question then is whatabout, What about future integration?
Maybe you have it now, but I'm guessingfuture integration with other real time
data that might prove handy to assimilateinto answers and so on, like ADS B data,
GPS data, um, data that's coming fromForeFlight or, you know, things like that.

(55:34):
Kind of, how are youtaking that into account?

Alex (55:38):
Right.
And so we're starting by using thisreally convenient interface that explained
just launched called the web socket

Bill (55:45):
Mm hmm.
Mm

Alex (55:46):
not have to install a plugin in order to be able
to have data go to Shirley.
And so that's been really very convenient.
We're looking at releasingan app version of Shirley.
You know, there are other, thereare other sort of platforms that
provide these sort of data streams.
So ADS B is a good example, butyou can have a data stream from
infinite flight, which is the iPad

Bill (56:07):
hmm.

Alex (56:07):
in.
Sort of one of the leaders there, youknow, you can incorporate surely that way.
Of course, in the cockpit, you can have ahave the GPS from the sky onto your phone.
But of course, you andthe altimeter actually.
Now there's also altimetersbuilt into into phones.
Believe it or not, there'sbarometric sensors.

(56:30):
But yes, the ADS B that comesin has a lot more reliable data.
You should use that.
One of the things that we've beenfocusing on as well, and one of the
reasons why we are sort of doingthis sort of data process is that, so
that surely will pay more attentionto the particular aviation context.
And so one of the things that we'venoticed is that existing really big sort
of language models are not very good atpaying attention to the state of flight.

(56:55):
And so even though you might haveinformation related to, of your altitude
and your descent rate and, you know, yourbank angle and things like this, you know,
the system might not fully be appreciatingthe fact that you're about to land or
you already have landed and insteadrelying more on the story to think about.

Bill (57:15):
Mm hmm.

Alex (57:16):
you've landed or not.
And so part of that is that thesesort of models like to reason,
sometimes you reason best by doingthis kind of chain of thought thing.
It's like, you know, you could tellit this, you could say something
like, okay, surely using theinformation, the information that
you see related to altitude andrate of descent, what would you say?

(57:39):
We're in the herd phase of flight.
And then it could saysomething along the lines of.
Descent rate is 500 feet per minute.
You know, altitude is 500 feet AGL.
We're about to land.
And, you know, what is really importantis to be able to skip those steps and just
for sure, to be able to understand that

Bill (57:56):
Yeah.

Alex (57:56):
about to land.

Bill (57:57):
Just

Alex (57:57):
And so bigger and better and badder models

Bill (58:00):
Yeah, right.

Alex (58:01):
able to appreciate those factors, but also by tuning the models using
that sort of information, you'reable to skip those steps and how

Bill (58:09):
Yeah.

Alex (58:09):
attention to the particulars.

Bill (58:11):
we've, we've got,

Alex (58:12):
opportunities to do this.

Bill (58:14):
sorry, go ahead.
Yeah.

Alex (58:16):
Well, there's other opportunities also to do things called multimodal
models for, you know, essentially1 channel is the flight data,

Bill (58:23):
Mm hmm.

Alex (58:24):
then the other channels are the speech.
And, you know, you could havetwo different channels feeding
the model so that it all points.
It has some concept of, you know, yourstate of flight that it's generating
and also some concept of the story and

Bill (58:38):
Right.
And that's kind of what I had in my mind.

Alex (58:40):
it into text land too.

Bill (58:42):
Right.
And that's sort of what I had inmy mind, because we have the, we
have access to these relativelyinexpensive adhars and things like
that, like in the century, right?
So it, that's kind of, is thatkind of part of your roadmap is
to have that multimodal approach.

Alex (58:58):
and that's 1 of the cool reasons for starting with Sims 2
is because they have all that same

Bill (59:03):
Got it.

Alex (59:04):
You're just getting it streamed from the flight

Bill (59:05):
Exactly.

Alex (59:06):
could absolutely trade to pay attention to this thing, especially
for a smaller model, which wouldhave a lot more trouble paying
attention to the correct thingswithout tons of prompting, et cetera.
And so being able to provide thatsort of these sort of pairs and being
able to train these certain models topay attention to their phase of life.
That's a, that's a huge

Bill (59:25):
It's very interesting.
And you mentioned,

Alex (59:26):
question.

Bill (59:27):
you mentioned these massive models that, you know, they're going to
get supermodels and things like that.
not the way I thought aboutthose supermodels in high school.
It's different.
But, um, so I thoughtyou'd like that, Tiffany.
Okay.
Um, so, so my question then is,I think you're iOS only, right?

(59:49):
Today?

Alex (59:51):
So right now we're X Plane

Bill (59:53):
Uh, explain only.
Okay.

Alex (59:55):
and so we're, so we're

Bill (59:56):
Running in the browser.
Okay.
Thank you.
Thank you.
so you're probably thinking about how thiswill, um, run in the cockpit and so on.
And so I'm wondering how importantthe, the differences that you're seeing
between, you know, the A chips in, ina, in an iPhone versus now, you know,

(01:00:17):
most of the iPads are getting M chips.
How important is that to you to beable to do this edge processing?
as you go forward, is that goingto be a big deal for you like
you're you'll be able to forkyour or or change your application
based on the power that it has.
If it's running on an iPad that hasan M four in it, it's going to be

(01:00:40):
able to do different things thanif it's running on a phone with
an alpha chip with an H up, right?

Alex (01:00:46):
Right.
No, I think that's, I thinkit's massively, I think
it's massively important.
In fact, this is only the first yearwhere you might have a mobile phone,
which is powerful enough in order tobe able to run an AI model on it at

Bill (01:00:59):
Right.

Alex (01:01:00):
And so, you know, the iPhone pros that exist right now can run AI
models that, you know, my iPad here.
Can can run, you know, the,the, there's two aspects.
One of them has been just sort ofmaking models a lot more efficient

Bill (01:01:13):
Mm hmm.

Alex (01:01:14):
packing more punch into these models.
So a really good one is one of thesmaller sort of these Facebook created
a medic rated as models called the llamaseries of models that are open source.
And so you can essentially, youknow, start to think about bringing
some of those things to mobile and.
You know, the, the opportunitythere is, is massive.

(01:01:34):
So you can run some of thesemodels already on iPads.
The of historical onesare really not that smart.
It kind of and confabulate is really thekind of the term of art related to that.
And then the ones that you cannow run on the iPhone will have
to be sort of tuned by us and withthe sort of data that we have.

(01:01:55):
That we collect in order tobe smart for the task, be able
to be relevant for the task.
And, you know, the idea isto be able to upgrade when
somebody has internet connection,

Bill (01:02:03):
Right, right.

Alex (01:02:04):
is really the fundamental.
And there's different ways of lookingat how to get internet connection.
And

Bill (01:02:10):
And presumably it.

Alex (01:02:12):
around,

Bill (01:02:12):
send learning data as well, once they have a connection.
Um, I'm guessing, so it'snot only going to enhance the

Alex (01:02:20):
to, it's important.

Bill (01:02:21):
Go ahead.
Sorry, we got a little lag.

Alex (01:02:23):
It's important to have feedback.
It's important to have.
No, that's all right.
It's important to have feedbackon whether or not what somebody
experienced was was good or not.
And so, you know, of course, youdon't want to be exfiltrating
all sorts of data while you're in

Bill (01:02:40):
Right.

Alex (01:02:41):
know, in fact, that's a lot of what the SIM community
is for, is to be able to Yeah.
Be able to, to be able to try out someof these things and, you know, be able
to ensure that these things are safe.
And so I think when I think about flightdata, I'm a lot more kind of, know,
a little bit more hesitant to thinkabout kind of like kind of sucking
that data off the phone, especially ifit's all running on somebody's phone

(01:03:04):
and without having to use anything on

Bill (01:03:06):
Yeah.

Alex (01:03:06):
on the cloud at all.
And, you know, but it is important tohave some sort of feedback mechanism.
So that really has to be treadvery kind of lightly, you know.
You know, I ultimately think about howto provide these sort of technologies
to, you know, airlines, you know,pilots unions, nobody wants to be able
to have that, like, listening to themand their livelihood is on the line.
It just, you know, it's

Bill (01:03:27):
Yeah, you guys, you guys have a lot of

Alex (01:03:29):
So,

Bill (01:03:30):
CS questions ahead of you to answer, don't you?
Heh heh heh heh.

Alex (01:03:36):
really fun

Bill (01:03:37):
Yeah.

Alex (01:03:38):
like.
A combination somehow now, as a, asa grown up now, I get to say that I'm
combining a lot of interest for theaviation with computer science you
know, if there's other people out therethat are see this as an interesting
opportunity, they should reach out,

Bill (01:03:53):
We're,

Alex (01:03:54):
if they're on these

Bill (01:03:55):
we're really in front of that hockey stick curve, aren't we?
I mean, we're just in theinfancy of all of this.

Alex (01:04:02):
I think that's what makes it such a fun time to be building things.
And, you know, our favorite thing is tobe able to work with customers directly.
And so, No, we have our discordchannel where it's really
exciting to see people chime in.
It's airplane dot team slashdiscord if you want to get on

Bill (01:04:18):
All right.

Alex (01:04:19):
but the, the people who join, you know, we end up interacting and, and, and,
you know, sometimes they're willing to hopon a phone call and give us some feedback.
Sometimes they're posting YouTubevideos about things and, you know, that
they experienced a Shirley, you know,flights through New York or, you know,
Different experiences were surely orkind of funny videos for, you know,
sinfluencers, which I think is an awesomesubcategory of the internet influencers.

(01:04:43):
It's amazing.
And so, so we learned so much from that.
And I think there's just the enthusiasmin this space to see people really
enjoying using their sins andseeing what's possible in sense.
I mean, there's a lot of reallyinteresting companies that are, that
are shipping things for For, for Simsnow, you know, you have like the ATC

(01:05:03):
land and you have sort of, you know,same tensions and beyond ATC, P2ATC,
which are using all this AI stuff tobe able to enable really good ATC.
You know, it's, it's.
amazing time to be kind of buildingand also being a flight simmer.
And, and so, yeah, it justcouldn't be, couldn't be any more
fun if I could think about it.

Bill (01:05:27):
All right.
Well, I, I've, um, exhausted the,the geek out portion on the product.
So I'm going to, um, I'm going to giveTiffany a chance to ask some of her
pent up questions that I know she's got.
And then I want to reallykind of geek out on the, uh,
aviation learning side of this.
I've got some questions about that too.
So go ahead,

Tiffany (01:05:46):
I think all my questions were answered.

Bill (01:05:48):
Oh, good.
All right.
All right.
Well, um, got a few then.
Um, we've talked a little bitabout how, uh, Shirley could be
used in the training environment.
I would love to hear your thoughts go alittle bit deeper into how Shirley would
integrate, uh, and I mean this in a nontechnical way, but integrate with a human

(01:06:14):
CFI to augment the training environment.
Uh, and enhance what they're doing witha, with a student or with a, a client
who's doing some advanced training,

Alex (01:06:25):
Yeah, that's a great question.
so we see, just to be clear again andreiterate, like, we see Shirley as being
a supplement to professional instruction.
You know, there's a portion of, uh,practice that surely can provide a lot of
value and it's actually pretty good, but,you know, there are so many conceptual

(01:06:47):
things that only a really a human can

Bill (01:06:50):
Yeah.

Alex (01:06:51):
can see.
And plus, there'll be things thatsurely does miss and, and, and.
There'll be opportunities toto enhance sort of, you know,
some of the some of the style.
And if the, you know, are also unknownunknowns related to some aspects of
this of this training, but, you know,what is really valuable about it is
being able to show up prepared foryour lesson and to be able to do fine.

(01:07:14):
You know, you're.
You know, you might have missed a weekwith your instructor due to weather
or somebody got And now you're likelooking at potentially needing to
repeat a lesson or you just went outand did some training with an instructor
and you had these deficiencies.
You need sort of this retraining or,you know, in a, you know, you started
with your PPL, you had these maneuvers,you know, you were a little bit, you're

(01:07:38):
kind of bad at terms about a point.
And now you're going to go, you know,you got your IFR, you're really like
book smart, you got your IFR real easy.
And now you're back in commercial land.
You got, it's on pylons and you'rejust like, Oh, goodness gracious.
And you're struggling.
And so, you know, maybe for somebody whois having an integrated sort of training

(01:07:59):
program that includes Shirley, then theycould have some sort of knowledge that.
Yes, actually, in PPL time, you werepretty deficient these sort of areas.
In fact, you might want to actuallyput some extra, extra energy into
them before you go out and startworking on these eights on pylons
or between your instrument or,

Bill (01:08:17):
Mm.

Alex (01:08:18):
know, after you get your instrument, make out an airplane
or do it or do it in the sim.
And so some level of integrationthat you would want to have, you
know, maybe between your, yourfour flight pilot sort of logbook.
You know, there's.
Some people who take thelogs real seriously inside
of the electronic log books.
Some people don't take them quite asseriously as, as the opportunity with

(01:08:42):
the actual paper log book, becausethere's a, there's a flight track and
things like that associated with it.
Electronic logbook.
Sometimes you feel like it might maybenot be, uh, might not be as but for,
uh, if that data is there, then itwould be really good to be able to grab

Bill (01:08:57):
hmm.

Alex (01:08:57):
able to provide that to Shirley as part of your sort of private pilot
sort of preparation and have, youknow, your pilot notes related to,
you know, what did you feel like youdid right and did wrong and what areas
that surely believe that you mightneed a little bit of more practice on.
Include those in sort of a logbook that'sintegrated between your real flying and
your virtual flying and be able to helppeople become a lot more prepared and then

(01:09:18):
to have, you know, instructors be ableto kind of be able to try this stuff out.
Maybe they already used it themselves,or maybe they have heard about it and
just think it's maybe worthwhile justto have, you know, if you're tired at
the end of the day, or like, and youand what your setup is, if you have like
a yoke and You know, you're a laptopand now you have to like go to your
desk in order to like plug, plug it in.
It's like, you're kind of tired.

(01:09:39):
It'd be nice to havesomebody like provide or

Bill (01:09:41):
Mm hmm.

Alex (01:09:42):
a little bit of extra to get you, to get you over that sort of line.
And so you sort of show up andShirley's like, like, let's go,
you're ready to do blah, blah, blah.
And you're like, all right, like, I'm, I'mjust there enough to be able to practice

Bill (01:09:56):
Yeah.
Mm hmm.

Alex (01:09:58):
like at 125 percent as usual.
And so with that in mind, there aredifferent sort of styles of training and
different styles of learning and teaching.
And I think one of the one of theinteresting things with Shirley is that
what people don't know is that they can.
You know, if they're not enjoying,or if it's not working for them in a

(01:10:18):
particular way, you can actually justtell Shirley to do it differently.
And Shirley will actuallydo something differently.
You can just ask for a differentsort of approach to working with you.
So, you know, there is someopportunity of Shirley's is getting
that last word in too much to belike, Hey, can you just like not do

Bill (01:10:38):
Just

Alex (01:10:38):
or mute so that, Oh, we're not doing this.

Bill (01:10:41):
from the old Saturday Night Live days.
You could, you could have herrespond appropriately to Simodana.
Simodana.

Alex (01:10:51):
Yeah.

Bill (01:10:52):
All right.

Alex (01:10:52):
Yeah, you

Bill (01:10:54):
so do you see,

Alex (01:10:55):
And

Bill (01:10:55):
do you, resist.
You might be a little too young for that,Alex, but you might not remember that.
Um, but do you, do you see,

Alex (01:11:03):
No,

Bill (01:11:03):
what's that?

Alex (01:11:06):
said me.
Not too

Bill (01:11:07):
Oh, yeah, yeah, yeah.
Um, do you see maybe a future wherethere could be a, almost a three way
collaboration between a professional CFI?

Alex (01:11:17):
Yes.

Bill (01:11:17):
the learner and surely where there's feedback going both ways, um,
kind of from what the learner may havedone on their own in solo practice, kind
of some of that feedback, making it tothe CFI to help the CFI understand where
to put emphasis and things like that.
Do you see that kind ofin the future as well?

Alex (01:11:40):
Right.
So in fact, I at Oshkosh this year,I met a company called Noble Flights,
and they make these really cool sort ofSR 22 SR sort of home flight sins that
are advanced aviation training devices.
So if you own

Bill (01:11:54):
Really?

Alex (01:11:56):
you could actually have it in your house and be able to have
sort of instrument proficiency.
You know, you have a, theyhave like a, it's like a 20
to 40 to 50, 000 piece of kit.

Bill (01:12:07):
Yeah.

Alex (01:12:07):
you know, it's, it's an

Bill (01:12:08):
Well, they're also flying a million dollar airplane.
So,

Tiffany (01:12:12):
Yeah.

Alex (01:12:13):
yeah, yeah.
Those are pretty expensive too.
But, you know, one of theinteresting things that they are
thinking about is how to providesome level of remote instruction.

Bill (01:12:20):
Oh yeah.

Alex (01:12:21):
so there is a really big, big opportunity.
In fact, there are some companiesright now that already provide.
You know, there's one flight simcoach that provides remote instruction
between that I don't endorse.
I don't, I'm not personally familiarwith them, but you know, it's, it is
a company that where you, uh, and Idon't not endorse it for the record.
It

Bill (01:12:39):
Yeah, yeah, right.

Alex (01:12:40):
company.
That's out there that provides

Bill (01:12:41):
You don't know enough about it.
So yes, I got it.

Alex (01:12:44):
know enough about it, but like the, um, but it's really, it's a really cool
concept where you have an instructor who'sable to join you in a simulation session.
Now if you combine that, a few thingsyou combine in here, you combine a, a
A AI sort of, uh, you provide, surelyprovide some level of instruction and then
you have the ability to kind of connectin an instructor to be able to provide

(01:13:07):
sort of the verification of how somebodyis actually doing in their progress.
And you sort of have a really neatcombination, the sort of thing that
we were talking about in the kindof commercial spectrum, which would
be really useful for them, butalso something that could provide
sort of a combination, you know.
You don't necessarily get homeand you don't necessarily have
something scheduled with somebody.
We want to practice.
And so surely is really good becauseyou don't really, you can just

(01:13:27):
sort of show up and truly is ready.
And then, you know, there's aopportunity, of course, to be able
to provide that sort of connection.
And so a lot of our technology thatis able a lot of our technology
that provides the sort of connectionwith Shirley and the individualist
technology that could be able toprovide that sort of session between.
An individual and another individual.
And that's something we're reallyexcited to be able to do, because,

(01:13:49):
you know, not and not just in, youknow, and to varying degrees as well.
So you'd be able to havetheoretically at at its extreme.
You think about to a logical conclusion,you have a VR headset that somebody
you're sitting inside the airplane.
You're able to interactwith all the controls.
You have an instructor toactually sit next to you in sort
of this virtual space inside of.

(01:14:11):
Your airplane and be able to sort of seeyou as you're doing these sort of things.
I think this would be this is sort of thelogical conclusion where things can go.
You send the VR headset to somebody.
Surely you can help get them preparedto be familiar with this airplane.
You have an instructor join themnext to them inside of the cockpit.
And be able to verify that theyknow how to operate this vehicle.
And so this is absolutely where this sortof our kind of our part of our ground game

(01:14:34):
vision of how we bring sort of advancedtraining and make it a lot more accessible
and representative to all sorts of flightbetween maneuvers to sort of flight
training device style, sort of instrumenttraining and et cetera, et cetera.
And.
And looking, and looking to do that aspart of supply, surely, So that's a,

(01:14:55):
that's a great and relevant question.

Bill (01:14:58):
Cool.
Um, so, um, as it pertains to teaching,uh, one of the, one of the things that
keeps coming up in my mind is that,um, this is super ambitious, right?

(01:15:21):
So where we met, Um, there's a,there's another company, you know,
doing a co pilot, um, sort of approachand we're actually going to be,

Alex (01:15:31):
yeah,

Bill (01:15:31):
Goose, we're going to actually be talking to them in a couple of weeks,
um, for, for the same reasons, this isall very interesting, um, what they're
doing is, is much more, um, would sayfocused on a very specific use case of
your co pilot in the cockpit, whereasyou're taking the approach of being sort

(01:15:54):
of, you know, A lot of different thingsto a pilot, to a simmer, to a pilot,
to a learner, to somebody who's flying.
It's very, very ambitious.
Um, how do you, do you balance what you'redoing with what you can deliver right now?
How, how's that going?
And, and how are you walking that line?

Alex (01:16:19):
I think that's an insightful question for two reasons.
One of them is that a, it's realisticto be able to only, it's, it's
realistic to say, Hey, you're lookingat doing this entire scope of things,
what can you actually cut off?
But I think there's another sortof conceptual reason why that's an
insightful question is that it'sactually challenging if you're a.

(01:16:42):
Individual hearing about what we're doing.
It could be challenging tothink, okay, how does this fit
into my life in particular?

Bill (01:16:47):
Um, yeah.

Alex (01:16:48):
you know, I think, yeah, and so I think there's, there's,
there's a few different ways.
And I'll try to get back to the question.
Also, of, you know, how are webalancing it long term as well?
And so the to the former question,the, the thing that we provide

(01:17:10):
right now is a copilot thatconnects into your flights in.
Okay.
And can, you can either fly withand sort of an open sort of setting.
It's not just a deadhead.
It flips switches.
It interacts with the SIM for you.
It can be sort of a.
Kind of a co pilot, it's your buddy,

Bill (01:17:28):
co pilot.
Yeah.

Alex (01:17:29):
pretty

Bill (01:17:29):
Right.
Yeah.

Alex (01:17:32):
And the other aspect that we're working on right now is
through our challenges system.
And so we have this sort ofconcept that we're building
off of, which is challenges.
Challenges have enabled us to buildsort of fun experiences from the Aaliyah
sort of training challenge to also ourentire kind of training curriculum.

(01:17:54):
And so what we're providing and whatwe're hoping to get feedback on in part
through conversations like this is tohave pilots try out this sort of private
pilot curriculum and see if it's meetingtheir needs and be able to get feedback.
And so, you know, it, it, it would be fairto say right now that surely is, In early

(01:18:19):
adopters product, so it is absolutelyfair to say that, you know, you try.
Surely it works.
The product works, but there arethat you can provide feedback on.
And then within a day or two,we go and we build it you
provide that sort of feedback.
And you have this sort ofiteration process, you know,
we're talking August 20th 2024.

(01:18:41):
Hopefully in a few months willbe through that sort of process
with the private pilot curriculum.
Just like we're starting to getthrough that process related
to some of the co pilot stuff

Bill (01:18:49):
Yeah.

Alex (01:18:50):
right now.
And, you know, we'll continue to do that.
You know, we're working with a companythat has a few hundred And so we'll be
getting feedback with them using ourproduct to see how they fly with Shirley.
A lot of those people arebecoming pilots as well.
we're working to develop thisproduct as a startup company.

(01:19:11):
There are your majorprerogative is to grow.

Bill (01:19:15):
Yes.

Alex (01:19:16):
so you kind of are like this little silkworm, you're like trying
to find light and you're like

Bill (01:19:22):
Yeah.

Alex (01:19:22):
kind of put energy into what works and like, maybe don't worry too
much about what's, you know, aboutthings that are not really working
and it's kind of like you water the

Bill (01:19:32):
Or the,

Alex (01:19:33):
growing as opposed to the

Bill (01:19:34):
yeah, you got to figure out a way to make revenue, right?

Alex (01:19:37):
garden.
That

Bill (01:19:38):
That's not always that clear.
It's not always that clear when you'redoing something conceptual like this.
It's like, okay, well, wheredoes the revenue come from?
It's, it's not that easyof a question to answer.

Alex (01:19:49):
And so what we ended up doing is we, we really closely
with the customers that we have.
And so we ended up having these reallyclose customer relationships where we
find out things that are important.
We watched the videos that they'reposting, if they're streamers.
know, and then we end up figuringout how to make their lives better.
Fundamentally trying to buildsomething that people want.
I think it is still very insightful tobe able to say that there are, there are

(01:20:16):
a huge number of things that this, thatthis could be, and there are a huge number
of things that we're building towards.
I think we're in an excellentposition to be able to start with
these really of large group ofenthusiastic home flight sim users,
help them bridge from sim to sky.
And.
Be able to provide training andintegrate with other sort of training
platforms and curriculum to help thembe kind of having a one stop shop that

(01:20:39):
could help them get into the cockpit.
And with the idea that people whotrained with Shirley would want to
fly with Shirley and the actual,an actual airplane because they
would have had this experience.
And so I think, I think thereis a path from sin to sky that
makes sense for a company.
It was sort of the data pipelinesand things like this that.
It's set up really nicely tomake this all sort of possible.
Yeah, but if you're a customer today,the question is, you know, if you

(01:21:02):
have, if you have explained August20th, 2024, if you have explained and.
You are interested in flying with Shirley,and you're interested in training,
and it's a great time to get involved.
You know, it's a great time tojoin the discord and interact.
And as we sort of build this out,and we're excited to have more
people to to work with and learnwhat to learn more about what, you

(01:21:24):
know, what customers want and sortof build something directly on that.
Yeah.

Bill (01:21:28):
Okay.
So I'm going to ask a really, um,
selfish, selfish question, but maybethis will extrapolate to, um, uh,
to many of your future customers.
Um, I think it will.
I'm kind of an everyman.
So

(01:21:49):
when, when you and I first startedtalking about how Fly with Shirley
could be used in a training environment,my, my brain started going to all
sorts of places that would help mespecifically and maybe to other people.
Like I said, um, when I'm studying,for a rating for, test for, you know,

(01:22:16):
for a check ride, an oral, whatever,or maybe I'm just, you know, doing some
instrument, um, you know, some approachesand I start to ask myself some questions.
I'll spend the next hour to twohours diving into the aim and
into the, the FARs and tryingto find answers to my questions.

(01:22:41):
So the first thing that poppedinto my mind is I would love to
have a resource that's so aviationspecific like this that I could ask
questions or have a conversation.
Is something like, um, off thetop of my head, um, what, what

(01:23:03):
does a, uh, what does a verticaldescent point look like on a chart?
And it can show me, or it might be,you know, you explain the differences
between these two approaches?
Or, um, remind me what the three differenttypes, maybe it's four, I don't know,

(01:23:26):
the three different types of night, uh,are, I, I'm trying, I'm trying to think
of examples, that's probably not thebest example, because that's pretty easy
to look up, but I'm trying to think ofexamples where you would want to get
specific with Shirley on things thatyou're not going to get from an internet,
internet search, because what you'regoing to get from an internet, internet.
An internet search is a bunch ofdifferent opinions, maybe a podcast,

(01:23:48):
maybe a YouTube video that's on thesubject, you know, that sort of thing.
But I might want Shirley to helpguide me in the research I would have
done anyway and help me do it faster.
So in other words, it, it pullstogether the answer, but then
points to the resources sothat I can quickly go the FAR.

(01:24:09):
you know, that specific section andsee in the AIM, the chart that they
have, you know, that sort of thing.
I hope I'm getting across what I'm asking.
Is, is that somethingyou're thinking about?

Alex (01:24:21):
Right.
And so, cool.
I think there are absolutely aspectsof that that we're thinking about, you
know, when you start to think about howShirley can provide a virtual check ride.

Bill (01:24:32):
Mm hmm.

Alex (01:24:32):
there are questions you'd want to be able to ask related to that.

Bill (01:24:36):
Yes.

Alex (01:24:37):
And then if the D.
P.
says, well, I actually don'treally know the answer.
You tell me that that doesn't reallysort of scratch the itch that you're

Bill (01:24:45):
Right.

Alex (01:24:46):
about here.
What I what I will say is that, youknow, plan to back up and surely with all
all the references from, you know, the.
You know, the, the far, of course,the firing and all these sort of

Bill (01:25:02):
Yeah.

Alex (01:25:02):
the handbooks, all the flying handbooks, the P.
O.
H.
is, et cetera.
There are, there are sort of like.
where you have sort of like FAAlegal department decision papers

Bill (01:25:15):
Yes.

Alex (01:25:15):
you know, what

Bill (01:25:16):
The legal opinions and stuff.
Yeah.

Alex (01:25:19):
who's, just like you get into that absolute wild land of, you know,
questions that I think are really goodquestions or the sort of questions
that I used to, Get crickets to and I'd

Bill (01:25:29):
Yep.

Alex (01:25:29):
on our sort of discussion forum be like, ah, who cares?
Like, why are you like, you know,or this is like such a specific
IFR question related to, you know,philosophers does not, you know,
these are the sort of like very.
Kind of insightful questions thatneed to be kind of built through,

Bill (01:25:47):
I've got some logging questions

Alex (01:25:49):
the,

Bill (01:25:49):
right now.
Here I am, um, I've been a private pilotfor 15 years or something like that.
And here I am at the commerciallevel and about to finish my CFI.
I was looking at my logbook the otherday, about ready to enter some stuff.
And I'm hitting situations whereI'm going, I don't actually

(01:26:09):
know the proper way to log this.
Like, it's not your typical, like, I'vegot a safety pilot situation or, uh, I,
I won't go into details, but my pointis I would love to have a resource
because I even know that most of myinstructors that I know are probably
going to go, yeah, I don't know.
it would be amazing to have aresource that would help you.

(01:26:31):
discover the answer for yourselfby walking you along a path.
Well, I don't know.
Good question.
Let's look at this.
Well, what about this?
And you know, it's, it would just beincredible to have a resource like that.
You know?

Alex (01:26:44):
I think that there are, I think that this could be in the
good wheelhouse for companiesthat are providing sort of ground

Bill (01:26:54):
Yeah, that's true.

Alex (01:26:56):
materials,

Bill (01:26:57):
point.
Yeah.

Alex (01:27:01):
able to give some sort of insights related to some of these
questions, you know, I think withthe thing with the format of Shirley
currently, you know, And as much aswe were just talking about how we were
intending surely to be everything to

Bill (01:27:14):
Right.
Right.
You've got to, you've got toput a limit somewhere, right?

Tiffany (01:27:18):
Yeah.

Alex (01:27:19):
we're, we're bridging from SIM to sky, but we're looking to basically
be the co pilot sort of setting.
So surely is your co pilot friendsitting next to you in the cockpit who
can also provide some level of trainingand instruction, not necessarily going
to like write up that has references tothings, although like I will say that,

(01:27:41):
you know, We'll come back to you and beable to get and get your feedback on that.
And, and, and.
In some time, let's say, you

Bill (01:27:48):
Yeah.
Yeah.

Alex (01:27:49):
let's say you are, you're, you're, you're saying probably resonates
with what a lot of other studentsexperience, which is, you know, really
stumper questions that are super hard.
You

Bill (01:28:00):
Yeah.

Alex (01:28:01):
fact, I've met some people who are working on these sort of written based
sort of like, you type it out, yourquestion, and then it's like providing.

Bill (01:28:09):
Oh yeah.

Alex (01:28:10):
of feedback.
I met somebody like thatat, actually at, at Oshkosh.
I wish I had his cardat the top of my mind.
again, another person thatI haven't tried there.
I mean, I barely tried their stuffand, you know, it seems cool.
We should check it outtype of deal, but the.
The, the, the solution there, Ithink, is, um, is a kind of thing that

(01:28:33):
spits out text that's permanent, thathas references, and is a very bulk

Bill (01:28:37):
Interesting.
Yeah.

Alex (01:28:38):
do

Bill (01:28:38):
Yeah.

Alex (01:28:38):
be part of a one stop, and that's a, and that is an event, that is a place
where AI can take a lot of advantage.
You know, what we'd like to be ableto do is be able to talk to you about
those sort of things, pull in therelevant information so that while
you're kind of having this discourse,that could be an important thing or
take a note for you to be able tolook that sort of thing up because you
might be flying an approach and you'relike, my goodness, what the heck is

Bill (01:29:00):
Yep.

Alex (01:29:01):
And,

Bill (01:29:01):
Yep.

Alex (01:29:02):
you know, you want to remember that later.
And so you'll be able to have that assort of a flight note that then ends
up in your, you know, inside of your

Bill (01:29:09):
Yeah.
Yeah.
Great idea.

Alex (01:29:11):
items

Bill (01:29:11):
Great idea.
Yeah.

Alex (01:29:13):
think it's really cool to be able to think about Shirley as being part of
a one stop shop related to some levelof ground, like ground instruction.
You have a groundinstruction integrated with.
A, um, you know, you have grounddestruction, really integrated with
private pilot maneuvers, integratedwith private pilot sort of checks

(01:29:34):
your check ride preparation.
And so, you know, I think it isconceivable that we could get
into building a whole learningmanagement system that provides a

Bill (01:29:43):
Yeah.

Alex (01:29:44):
This sort of text based thing, but we're not doing

Bill (01:29:46):
Right.
Right.
Right.
Right.

Alex (01:29:48):
you know, I think, that needs to be seen.
So I will be the exception tosay that we're not saying that
we're everything to everybody,

Bill (01:29:54):
Awesome.

Alex (01:29:55):
that's my

Bill (01:29:56):
Look, we came full, we came full circle on that.
Look at that.

Alex (01:30:00):
That's right.
Yeah.

Tiffany (01:30:01):
Did you say when, what year you started this?
Was it?
I see March 24 on your LinkedIn.

Alex (01:30:10):
Yes.
March, 2024.

Bill (01:30:11):
It's brand new.
They're, they're at theirfirst MVP basically.
Well, a little past it now, but.

Alex (01:30:17):
So flights in Mexico was our was our MVP.
That was in July.
so at the beginning of July, um,and so, yeah, we've just been

Tiffany (01:30:28):
Wow.

Alex (01:30:29):
a couple months now.
And so, yeah, anyway,

Bill (01:30:32):
Excellent.
Well,

Alex (01:30:33):
exciting.

Bill (01:30:34):
it is very

Tiffany (01:30:35):
it's

Bill (01:30:36):
and it's, it's, It's hard to comprehend sometimes where all of this
can go, uh, but we certainly appreciateyour time helping bend our brains a little
bit around, around the possibilities.
And because I know you've been thinkingabout this a lot more than we have.
So we do

Tiffany (01:30:56):
well, and I

Alex (01:30:58):
been my

Bill (01:30:59):
time.

Tiffany (01:31:01):
I could be wrong.
I know we at least used to hang outwith the, the ForeFlight guys or they
were kind of in that circle and it wasthis tiny little startup and like no
one's ever going to let you use yourphone in the closet and look at it now.

Bill (01:31:15):
And look at

Tiffany (01:31:15):
this is, it's really exciting to see this.

Bill (01:31:17):
Well, and they're, and they're starting to use, I mean, it's a
totally different thing, but they'restarting to use big data as well.
Um, you know, they're, they'repulling data from those adhars
in the centuries to be able to doautomatic pyreps on turbulence.
You know, and so they're using,they're using models to filter out,

(01:31:43):
operations so that they can seewhere, um, turbulence is happening.
They know the airplane, youknow, they know the type, they
know the weight of the airplane.
They know exactly where it's at.
They know what altitude it's at.
They know what direction it's heading.
they're filtering out normalmovement, normal airplane movement.

(01:32:05):
And what they're left with is Thatwas turbulence, and that, this
is the level of that turbulence.
That's, that's pretty incredible stuff.
I mean, it's totally different thanwhat you're doing, but the idea of being
able to take, um, large amounts of datafrom a lot of different places and,
and give real time insights to usersof the application, it's really amazing

(01:32:26):
what, what we're going to be able to doover the next, you know, 10, 15 years.

Alex (01:32:33):
I'm going to say that there are so many.
You know, the, these companies thatstarted in the last generation of
product, whereas essentially, you know,we have these new iPhones and iPads,
how can we bring charts to them and thenalso make pilots lives a lot better?
Those things have gotten so good.
I

Bill (01:32:51):
Yep.

Alex (01:32:52):
look at ForeFlight, it's just an amazing piece of technology.
It has all the datapossible on an entire world.
And now what's really neat aboutthis sort of technological transition
is that there are all thesereally cool companies that exist.
let's sort of look aboutthis next generation.
So you have, you know,if you have goose is a

Bill (01:33:08):
Mm hmm.

Alex (01:33:09):
of that.
You have sort of the same intentions.
be on to see, you know, you have,you know, you have surely in this
sort of category of being able toprovide training and eventually be
able to get back into a cockpit.
Something that's pretty conversational.
You have companies like.
You know, loft dynamics, which aremaking sort of, you know, VR certified
Sims or maybe 30, like divide by 20, the

Tiffany (01:33:30):
A

Bill (01:33:30):
right.

Alex (01:33:32):
Sims.
And, you know, it'sjust an incredible time.
I would say to be a, a pilot,a simmer, a flight school, just
so many things are gonna be a

Tiffany (01:33:42):
startup?

Alex (01:33:44):
and,

Bill (01:33:44):
A startup

Alex (01:33:45):
and a

Bill (01:33:45):
let let's face it though.
It's always a great time to be a pilot.

Tiffany (01:33:51):
It's true.

Alex (01:33:51):
It's always a good time to be a pilot.
Yeah,

Bill (01:33:54):
couldn't resist.
Excellent.
Oh, this is, this

Alex (01:33:58):
good to have.

Bill (01:33:59):
has been really fun and really, really enlightening.
So Alex, wonderful time.
We appreciate it.
We've held you long enough.
Um, so we'll let you go.
Anything that we forgot to ask ordidn't ask that we should have.
You want to make sure we cover

Alex (01:34:14):
it just, well, it just encouraged folks that they can,
you know, follow along with whatwe're up to over at airplane.
team slash discord or slash blog.
If they're curious, youknow, we, we love pilots.
We love Simmers.
talking to our customers, you know,it's, it's, It's real important

(01:34:35):
to us that pilots and aviators andinstructors continue to We want to
see more people becoming pilots.
We really care about this entire industry.
I love flying and, you know, justwant to wish everybody to fly safe.
It's been really good talkingto you, Bill and Tiffany.

Bill (01:34:52):
beautiful.

Tiffany (01:34:52):
awesome.
Thank you so much, Alex.
This was great.
We look forward to seeingyou at Oshkosh next year.
Are you going to have a booth?

Alex (01:34:59):
Likewise.
I think we'd like to probably figureout something like that the next

Bill (01:35:05):
Yeah.
Excellent.

Alex (01:35:06):
it's probably going to be a good time.

Tiffany (01:35:07):
I think so.

Bill (01:35:09):
Well, thanks again.

All Episodes

SPC#79 - And Don't Call Me Shirley

Episode Transcript

Popular Podcasts

Stuff You Should Know

Dateline NBC

The Burden

.css-15opob5{left:0;position:absolute;top:0.8rem;} All Episodes

.css-14f5ked{margin:0;word-break:break-word;display:-webkit-box;-webkit-box-orient:vertical;box-orient:vertical;-webkit-line-clamp:2;overflow:hidden;}SPC#79 - And Don't Call Me Shirley