May 14, 2025 • 29 mins
Hudson Fuller, GK Technology's newest full-time employee and computer science graduate, shares his journey from baseball player to agricultural data specialist working on the ADAPT standard integration.
• Hudson's connection to GK began through family ties and company retreats before pursuing internships during college
• Completed a computer science degree at Dakota State University while playing college baseball
• Created an AI baseball pitch predictor that achieved 80% accuracy in predicting pitch outcomes
• Currently coaches high school baseball and plays for the local amateur team
• Working on implementing the ADAPT standard to unify agricultural data formats across different systems in ADMS
• ADAPT aims to standardize how agricultural data is stored and transferred between different controllers and software
• The standardization is similar to how hydraulic hookups were standardized to allow equipment compatibility
• Different file types (shapefiles, GeoJSON, ISO XML) serve different purposes in storing agricultural data
• Modern controllers can process thousands of data points instantly to vary application rates accurately
Join us for part two of our conversation with Hudson in the next episode of Ag Geek Speak.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Sarah (00:00):
And now it's time for a Geek Speak with GK Technology's,
Sarah and Jodi, friends and Ican't wait to get in the fields
again.
Hudson Fuller (00:17):
No, I can't wait to get in the fields again.
Jodi (00:31):
Welcome back to Ag Geek Speak and welcome to a very
fantastic episode.
We have a very special guestwith us today, hudson Fuller,
who, as of right now, beginningof May, is our newest full-time
employee, so we're super excitedto have on the podcast to talk
about his background.
He is a recent graduate incomputer science and I'll let
him talk more about hisbackground there and also
(00:53):
someone that's working veryclosely with integrating ADAPT
standard into ADMS, and Hudson'sgoing to give us a lowdown on
that and talk more about how weget data from our ag equipment
into our computer.
So we're really looking forwardto this episode and we know
that Hudson will have a lot offantastic information for all of
(01:14):
you listening to the podcast.
So with that, hudson, do youwant to give a little bit about
where you're from, how you mayhave become familiar with GK
technology in the first place,because for those very keen
observers, hudson's last namemay sound a little familiar to
those that have worked for GKfor a while.
Hudson Fuller (01:34):
Yeah, so I'm Hudson Fuller, as Jodi stated,
and I'm from Clark, south Dakota, and I was first introduced to
GK at the first company retreatI got to go to.
As we're a very family orientedcompany and the kids get to come
along and my dad works for GK.
That was really the first timeI got integrated in with Darren
(01:56):
and Kelly and got introduced tothem.
And then got to my sophomoreyear of college and thing I was
looking around for internshipsin the area and I figured I
might as well talk to Darrenbecause I knew I enjoyed the
outdoors and agriculture fitright in with that.
But I also loved the, the nerdstuff with the computer science.
(02:17):
And I talked to Darren and hewas willing to give me an
internship for that summer wheremy main focus was working on
the first version of Adapt.
And from there, fast forward tothe next summer I worked with
GK again as an intern, alsoworking on Adapt.
However, it was a newer versionof Adapt where they completely
(02:37):
rebranded everything and then,following that, they decided to
give me a full-time offer inDecember.
I accepted it and started towork here in January.
Sarah (02:48):
So wow, there's a lot there, hudson, that's a lot.
So my first question for you ishow old were you when you went
to your first GK retreat, the GKTechnology Inc.
We go on a company retreatevery summer.
It's about a week long, it'sour company meetings for the
(03:09):
year, annual meetings, where wereally get a chance to like hash
through everything for the yearand where we're going to go for
the future.
And you know, as Hudson said,that's when family gets to come
in.
But how old were you on yourfirst company retreat?
Hudson Fuller (03:23):
I was 17.
Yep, some of them have beenthere a lot longer than I have.
It just never worked out thefirst couple of years of the
retreat for me to get to go.
But I got to go 17, right aftergraduating high school.
Sarah (03:37):
That's awesome.
Okay, and what was yourfavorite thing that you got to
do for fun on that retreat?
Hudson Fuller (03:43):
It was probably.
It was in Montana, up in theBillings area and we went to a
natural land bridge and myfavorite.
That was my favoritedestination that we went and did
and it was cool because I wasable to throw our rock across
the land bridge, get it acrossthe whole canyon.
Jodi (04:01):
I like that reasoning.
Sarah (04:03):
That's probably because you might know a little
something about baseball andthrowing a baseball.
Yep, I like that reasoning.
That's probably because youmight know a little something
about baseball and throwing abaseball.
Hudson Fuller (04:08):
Yep, I played college baseball for three years
and played throughout highschool and everything else at
Dakota State University.
Sarah (04:15):
Awesome.
So, speaking of Dakota StateUniversity, what did you major
in?
I don't think my computerscience description was really
encompassing of what your majorwas.
What did you major in, and wasthat first what you started in?
Tell us more.
Hudson Fuller (04:30):
Yeah, so my major was computer science actually.
Yep, you got it right and thenno, that's what I started with.
At first I was pretty sure Iliked computers enough to do it,
but once I got into some of theclasses where you actually were
hands-on doing stuff withcomputers, I realized that I
actually enjoyed doing it a lot.
(04:50):
It didn't feel like I was inschool, it just felt like I was
learning something for fun.
Sarah (04:54):
We are certainly glad that we have you around here to
solve all of our programmingissues, and you think it's fun.
That's great.
Jodi (05:03):
So what does that encompass?
Do you learn a little bit aboutcoding, about the hardware
Maybe?
What was your favorite classgoing through the whole computer
science degree program?
What was your favorite specificclass out of that?
Hudson Fuller (05:17):
Yeah.
So going through it was prettymuch all about coding and then
the theory behind coding.
It was pretty much all aboutcoding and then the theory
behind coding.
Some classes were every weekyou'd have some sort of coding
assignment where you'd have tomake something.
Other classes it was more aboutdo you understand how this idea
works, writing it out on apiece of paper.
(05:38):
And then, as far as my favoriteclass, it was probably in my
last semester, it was a softwareengineering class because it
was all about one big groupproject, basically the whole
semester, and it was my firstreal introduction to group work
in a code system.
But my two internships with GKI was kind of off on my little
(06:01):
lonesome working on the adaptbecause no one else understood C
sharp, which is the programminglanguage that adapt needs.
So it was awesome to be able towork through different pieces
at once to come together to afinal.
So it was an AI baseball pitchpredictor.
(06:32):
So like the purpose of it was topredict which pitch you should
use to trick the batter, or soit was more about say you knew
you were going to throw a sliderand then you could figure out,
based on how hard it was whereit was at in the strike zone and
what the count was.
(06:53):
It would figure out what themost likely outcome of that
pitch was, whether it wasgetting put in play a strike or
a ball.
Sarah (07:01):
So you could predict the strikes or the balls based on
what pitch you were going tothrow.
Hudson Fuller (07:06):
Yes, okay.
Jodi (07:08):
How accurate was it at predicting at the end?
Hudson Fuller (07:12):
It was between those three outcomes, it was, I
think, 80%, which is quite goodfor baseball, Because baseball
there's a lot of things that gointo baseball and lucky things
will happen that you can neverpredict.
Jodi (07:28):
That's really interesting.
So you kind of need a databaseto do that right.
Like you're kind of buildinglike a regression model and like
predicting response.
Where did you, where'd you pulldata from and like, were there
any factors that like came outthat were really predictive,
that you didn't think they'd be?
Or were you kind of limited on,like what you were pulling in
for?
Hudson Fuller (07:48):
Yeah, so there is a website called Baseball
Reference and it is for thenerds out there.
It has everything from how mucha pitch moves, exactly how hard
it is, to how many times perminute it's spinning, and there
is a library in Python thatallows you to pull all of this
(08:09):
data just almost instantly.
And you can pull back to, Ithink, it's 2012.
You could pull every pitch fromthat season, from a certain
team, from a certain player.
There's so many different waysthat you could pull in the data,
and that's where we struggledat first, because we could
export it in an Excelspreadsheet just to see it,
(08:30):
instead of trying to figure outby coding it what is actually in
there.
We thought that how much apitch moved was going to have
way more of an impact than itdid.
It ended up having almost azero impact.
It was way more about where inthe strike zone that pitch
actually was.
Sarah (08:49):
That's so cool, and did I hear you correctly a little bit
earlier that you just calledyourself a nerd?
Hudson Fuller (08:56):
Yes.
Sarah (08:57):
That's awesome.
I mean, we at GK actually liketo classify ourselves as geeks,
but that's okay.
I have also been known to beaccused of being a nerd as well,
so we're all in good companyhere.
That's awesome.
Hudson Fuller (09:12):
Nerds make the world go around.
Sarah (09:15):
So do geeks.
And here we are.
Jodi (09:19):
I love it because it's very similar to putting together
a science project.
Right you have a hypothesis ofwhat you think is going to
affect the outcome, right.
And then you test it.
In this case, you brought thedata back, you looked at it, you
built a model and found outthat it was actually a different
factor that influences it themost.
Sarah (09:35):
But on top of that is where you were talking about how
there's a lot of data that'sout there and, you know,
(09:57):
sometimes in baseball there'ssome things that are just
unpredictable and you just can'tpredict it.
That really isn't thatdifferent from agriculture.
There's a lot of things that weknow, there's a lot of things
that we deal with every day inthe field and we base a lot of
(10:21):
our decisions based in previousknowledge, in data that already
exists.
But sometimes there's thingsthat just happen, right yeah.
Jodi (10:26):
That's a great point, Sarah, Because like, okay, so
you've got all this data right,and I think we've been talking
in agriculture for the past oh Idon't know how many years ever
since we've gotten yieldmonitors on combines.
But we have a lot of data inagriculture now and it's kind of
piecing through and figuringout what actually has value and
what doesn't.
And that's kind of excitingwith AI right, because instead
(10:53):
of having to choose what factorsor like what data you're
pulling in to build your models,you can now kind of bring
everything in, as long as youhave a library that's organized
and I think we'll talk aboutthat with Adapt but you can kind
of look at all of it and figureout how that might predict
final yield at the end of theseason, and maybe that's a
little bit too optimistic atthis point in time, but that's
cool.
I think that's where the futureis is using artificial
(11:13):
intelligence to analyze a lot ofthe stuff that we have.
Sarah (11:15):
So lots of talk about baseball and data and how we're
doing predicting what's going onwith baseball.
So, but you've been reallyactive with baseball, so what
tell us a little bit about, likewhat exactly your history with
baseball has been?
Hudson Fuller (11:29):
Yeah.
So the first memory any of myfamily and me have with baseball
is me wanting to play t-ballwhen I was three, and we
couldn't until I was four, and Iwas so upset at that point in
time.
It was one of my sister'sfriends that was coaching that
team and it was hard for themnot to let me play.
But I had to wait one moresummer I did, and then worked my
(11:53):
way up from t-ball to coachpitch and then the first year
where I ever pitched at allwould have been probably when I
was 12, and I only think I threwtwice and it was something the
previous coach wanted me topitch more, but I didn't want to
, and then it just never reallycame up much until I.
It got mentioned once that I'dlike to try it, so we tried it.
(12:17):
It was okay, but it wasn'tfantastic.
Then we get up to the big fieldthe size that every team in
America plays once you're about14.
And then it was.
I got thrown on the moundagainst a bunch of 18 year olds
when I think I was 13.
And my second time ever doingit I had a line drive come back
(12:37):
and hit me in the wrist and itwas that we went to the
emergency room because wefigured it was going to be
broken.
It hit me that hard, Jeez Yep.
Then kept working through highschool and then I think it was
about my junior year I realizedthere was a chance I could play
college baseball and I wasreally interested in doing it.
So I emailed the coaches atDakota State because I knew
(13:00):
there was a chance that I wantedto do computer science and the
DSU has the best computerscience program in the area.
So I was like it would makeperfect sense if I would be able
to go play baseball there.
I got an email back sayingwe'll talk once you actually
turn into a senior, Got an emailback and went down for a visit
(13:21):
and got my scholarship offerthere.
Sarah (13:24):
Oh, that's awesome.
So you got through collegeplaying baseball.
Yes, that's really cool, that'spretty neat, that's that's
pretty cool.
So, and are you still?
Is baseball still a part ofyour life today?
Hudson Fuller (13:40):
Yes, so I actually coached the high school
team and the 16 year old teamhere in Clark.
It's awesome to be able to dothat and give back to the.
I've played with some of thekids too, because we've had to
play 13 year olds with the 18year olds, and then I also play
for our local amateur team inthe summers okay, so that, so
(14:01):
baseball still still a big partof your life and everything else
so it was.
Sarah (14:07):
It was always kind of interesting because, uh, I know
your parents followed you arounddoing taller games and
everything.
And it was always kind of funbecause I think Paul, your dad,
is Paul and he used to take offand he'd bring his computer with
and he'd map, you know, on theroad going all the way to see
you and everything like that andget a chance to catch up on the
(14:29):
baseball and then come on backagain.
So it's, it's interesting.
We've been hearing about Hudsonand his baseball as as
employees of of GK for for alittle bit.
Well, that's pretty neat.
I think Jody was asking acouple of questions about ADAPT.
So your first internship withGK was ADAPT and your second
(14:51):
internship was ADAPT, but alldifferent ADAPT, correct?
Hudson Fuller (14:55):
Yes.
Sarah (14:56):
Why don't you tell us a little bit about that?
What is ADAPT?
What does it do?
Yep.
Hudson Fuller (15:00):
Everything else.
So ADAPT is a standard thatAgGateway has set in an attempt
to unionize data formats acrossall of the agricultural software
companies.
So everyone can say, say, putit into a John Deere monitor and
pull it back out, and they'llbe able to turn it back into the
data they want to use in theirown software system.
(15:23):
The goal of it has always beenfull unionization and a lot of
different small companies havesigned on to make it a
possibility.
There are not that many thatactually have ADAPT working at
all and it's because it keepschanging, which is a great thing
, because it needs to keepupdating.
(15:46):
So the first version of ADAPTwe'll call the ADAPT framework.
That is the one that I think itwas.
Three or four years ago theidea was brought on to make a
standardization and, darren, heknew about it, he wanted to
implement it.
He just did not understand thecoding language that it ran with
and then it happened to workout perfectly that I did for my
(16:09):
first internship.
So we worked through that firstsummer and with that, with the
ADAPT framework, it could exportinto multiple different file
types, that being GeoJSONs, iso,xmls and then also an ADAPT
file, itself known as an ADF.
Sarah (16:27):
Okay.
So when you're getting andthat's the as applied data, then
pretty much that you're talkingabout coming from, okay, from
controllers, getting into astandardized format that we can
just bring in as generic fileformats.
Hudson Fuller (16:41):
Yes, yeah.
And then from there we got thatsomewhat working at the end of
the summer.
However, there was a few issueswith it with it.
And then we realized that adapthad changed something in their
code that made it not securelysigned, meaning we could not
release it to the public.
And then come to the nextsummer we know we figure out how
(17:06):
to fix that, I get that working.
And then we realized that adaptchanged everything.
They decided to get rid of thedifferent types of files that
Adapt would export to, and justalways export to, a JSON file,
which is it's not a bad idea atall, because then you don't have
(17:26):
a whole bunch of different filetypes floating around.
But that meant I had to recodeeverything so we could export
into that JSON style.
Jodi (17:36):
So, backing up even before Adapt, can you tell us a little
bit more about what the problemwas and why is there this drive
to have something standard?
And I think in my head I knowsometimes we get asked as
employees at GK to handledifferent information types and
I know sometimes I just can'topen them right away or like I
don't have the right import.
(17:56):
Can you tell us more about,like, what the problem at hand
is here and then how ADAPT andthat framework is helping to get
us closer to being able to viewand use this data?
Hudson Fuller (18:10):
Yeah.
So it's as simple as instead ofhaving a whole bunch of
different fruits and trying tomake a big fruit salad, you just
have one type of fruit set outat a buffet.
There is no real point in anyindustry, but in the agriculture
industry it's been waydifferent, because there are
more file types.
There's the shapes and everysingle different file type
(18:32):
associated with the shapes.
There's the shapes and everysingle different file type
associated with the shapes.
Every single one of those justleads to more space taken up on
a computer and leads to morethings having to get put on a
flash drive or whatever else toget put into a monitor.
If everything could get put inas one, our space would be a lot
(18:57):
smaller and there would be noconfusion within the industry.
Everything, everything fallsunder just a few file types,
instead of 50 or 100 that mightbe out there even though we
don't use them.
Sarah (19:02):
There are so many out there that are not used in our
area of the world but might bein australia, asia or anywhere
else so kind of like hydraulichookups back in the day then,
like John Deere had their ownspecialized hydraulic hookups
for many, many, many years andyou couldn't hook up like if it
(19:25):
was a chisel plow or somethingfrom a different brand.
You couldn't hook it up to theJohn Deere unless you had the
converter to convert thehydraulic hookup.
Hudson Fuller (19:36):
Yes, that's a perfect analogy.
Sarah (19:39):
Once we standardized hydraulic hookups, now we could
hook up whatever equipment wewanted to whatever tractor we
wanted, without converters,because we all had the same
hookups.
Hudson Fuller (19:51):
Yes.
Sarah (19:52):
Every color would talk to every color.
Hudson Fuller (19:54):
Yeah, that is a perfect analogy for that.
Jodi (19:58):
Would this get us to be where, like we wouldn't have to
use the FOD importer for yielddata that kind of thing anymore,
or I suppose, like there'dstill be older controllers that
would export like that correct?
Would this be something thatcomes out in newer controllers
and we'd see that newercontrollers are using the ADAPT
standard, or is this like astandalone?
How does ADAPT get implemented,I guess would be my question.
(20:21):
What do things look like onceADAPT is implemented?
Hudson Fuller (20:27):
Yeah, so the idea is to not truly replace FOD
because the older monitors andcontrollers they will never get
updated to ADAPT Okay.
But a lot of the newer ones Iknow specifically with Case they
have the full implementation ofADAPT within their newer
monitors.
I don't know the numbers thatcontroller monitors are, but I
(20:50):
know they use ADAPT and that'swhy we've been pushing hard to
get the ADAPT framework out,because they use the ISO XML
plugin of Adapt so that itexports into a task data into
that style format.
But it all uses a format insidebuilt from Adapt.
Sarah (21:09):
So would we be exporting out a geo JSON then essentially
the ADAPT system that the CASEcontroller could read.
Hudson Fuller (21:17):
CASE uses an ISO XML file that has the boundary
and it has the averages andcertain other numbers associated
with your prescriptions, andthen it also exports a bin file,
and that bin file has a gridassociated with all of your
(21:41):
different prescriptions forvariable rate for that field.
Jodi (21:44):
That's really interesting.
I guess I'm saying it'sinteresting because I've never
thought about you know, becauseI think right now for exporting
to cases you know, or likeimporting yield data, it comes
back in like a CN1 file or likea.
My point is like, even with likeKMLs too, when anything is
(22:04):
different from a shapefile, Idon't know how it works right,
like the thought of like havinga bin that has the grid of like
all of the prescriptions, itjust kind of blowing my mind.
Right now I'm just trying tothink through, like what are
could you talk about like someother types of like file types
and like how that's set up?
(22:25):
Like could you walk through,like in your computer language,
computer science knowledge, canyou talk about like what is a
shape file?
And then like what does a caseprescription file look like?
And maybe like another exporttype that you can think of?
Because I think when I knowwhat these things kind of act
like, it helps me reallyunderstand, like how to export
(22:47):
things better out of ad master,like why I need to jump through
a certain hoop to get what Ineed to get done yeah, so we'll
start off with a shapefile andwhen I'm actually working with
the data from a shapefile Idon't look at the shapefile.
Hudson Fuller (23:00):
I'm not seeing the boundary or the polygon
style prescription.
I'm not seeing that at all.
I am just seeing the data inthe back end.
So we have a tool within ADMSthat allows us to read in all of
those points on the shapefile.
Usually it is in differentfeatures.
(23:21):
That is like in the backendwhat a shapefile has.
Each polygon is not a polygon,it is a feature and each of
those features has so manydifferent values associated with
it.
If it's relating to aprescription, they will have
some sort of numerical valuevalue say it's 112 pounds per
acre, whatever it may be andthen they'll also have an extent
(23:45):
, is what we call it, and thatis the edge of that feature.
So every little square has fournumbers and that is the lat
long of each corner of that.
And then in ADMS we always readit in as northern and easting
in the UTM style.
Sometimes we have tore-transform it back into lat
(24:09):
long to make it so that we canuse that data across the world,
not needing to know the UTM zone.
That specifically has come uprecently when I've been working
with the field info stuff andtrying to figure out how can we
best actually get where thisfield is.
But that's a different topicfor a different time.
(24:30):
And then the Adapt ISO XML files.
When you open up that file, theISO file, you don't actually
see a picture of the boundary.
You will see probably a hundredpoints or so that make up the
edge of that field's boundary.
Along with that you will havethe average of the prescription,
(24:54):
like what the average poundsper acre in that field is.
You will have a default valueand a out-of-field value.
These values are in there.
So if for some reason thecontroller cannot use the bin
file, which is a file that justyou can't actually see the data
in there, the computer can seethe data though it knows exactly
(25:17):
what each point in that fieldthe prescription can be.
And then really another thirdtype that gets used is a GeoJSON
, which we also, through ADAPT,will have the ability to export
a prescription from.
That comes out as two separateGeoJSON files in one folder.
(25:38):
The one contains the basicinformation, such as the
boundary, the default rates, theout-of-field rates, and the
other, geojson, will be a very,very long JSON file that has a
whole bunch of differentboundary points in it or
prescription points, and then itwill have the variable rate for
(25:58):
each point within that field.
And, for example, one of thetest fields that we have through
GK, the Bloom House one, thatone ends up at 25,000 different
points of information that endup getting thrown into that
GeoJSON file.
Sarah (26:15):
Wow.
Hudson Fuller (26:16):
But the controllers that read that can
do it in an instant.
Jodi (26:40):
Wow that the controllers that read that can do it in an
instant.
When you increase theresolution or like decrease it
down, where you're going fromlike three meters to two meters
or like 10 meters to sevenmeters, you're increasing the
number of data points that needto be collected for each of
those, the squares, and Isuppose, for like older monitors
, like we see in, like themanual for exporting, um, that
that matrix, right, some of theolder controllers can't handle
that much data, right, right,and we've got to reduce that
(27:01):
polygon size.
But that makes sense If we'vegot data at all those points,
that takes up more space andmore memory with the computer to
read those and process whatthat actually means.
To tell the spreader or theplanter in the back what to do
and how to vary the rate, thespreader, the planter in the
back, what to do and how to varythe rate.
Sarah (27:21):
Gosh, I'm so glad that you're talking with us about
this because you know, from ourperspective, or like for me,
especially as agronomists, I seea lot of these things on the
front end but I've got no ideahow it's actually working in the
computer to see these thingsand convert it into something
(27:41):
that is, you know making ratesvary in the back, so this is
really fascinating for me tolearn from you, hudson, and not
only making the rates vary, butmaking the rates vary in the way
that makes you know theagronomist has the rates set up,
but we want to keep theintegrity of that prescription,
and not every single exportprocess can do that well, and so
(28:04):
it's really exciting to hearabout how this all comes
together, and not only that, butto have the explanation of all
of those different file types.
That's pretty fun.
Jodi (28:15):
So we're going to cut it here, but we've had such a
fantastic conversation justgetting to know Hudson a little
bit better, learning about hisbackground and how baseball has
helped make Hudson a bettercomputer programmer.
But we've also talked a littlebit about the ADAPT standard and
some of the things thatHudson's working on, and it's
been such a pleasure to haveHudson explain these things to
(28:36):
us and I know that we're goingto do the same when we come back
with Hudson in the next episode, in the second part of our talk
with him.
So thank you all so much forjoining us today and we will see
you on the next full-lengthepisode of Ag Geek.
Speak with GK Technology.
Sarah (28:53):
And at GK Technology we have a map and an app for that.
And now it's time for a Geek Speak with GK Technology's,
Sarah and Jodi, friends and Ican't wait to get in the fields
again.
Hudson Fuller (00:17):
No, I can't wait to get in the fields again.
Jodi (00:31):
Welcome back to Ag Geek Speak and welcome to a very
fantastic episode.
We have a very special guestwith us today, hudson Fuller,
who, as of right now, beginningof May, is our newest full-time
employee, so we're super excitedto have on the podcast to talk
about his background.
He is a recent graduate incomputer science and I'll let
him talk more about hisbackground there and also
(00:53):
someone that's working veryclosely with integrating ADAPT
standard into ADMS, and Hudson'sgoing to give us a lowdown on
that and talk more about how weget data from our ag equipment
into our computer.
So we're really looking forwardto this episode and we know
that Hudson will have a lot offantastic information for all of
(01:14):
you listening to the podcast.
So with that, hudson, do youwant to give a little bit about
where you're from, how you mayhave become familiar with GK
technology in the first place,because for those very keen
observers, hudson's last namemay sound a little familiar to
those that have worked for GKfor a while.
Hudson Fuller (01:34):
Yeah, so I'm Hudson Fuller, as Jodi stated,
and I'm from Clark, south Dakota, and I was first introduced to
GK at the first company retreatI got to go to.
As we're a very family orientedcompany and the kids get to come
along and my dad works for GK.
That was really the first timeI got integrated in with Darren
(01:56):
and Kelly and got introduced tothem.
And then got to my sophomoreyear of college and thing I was
looking around for internshipsin the area and I figured I
might as well talk to Darrenbecause I knew I enjoyed the
outdoors and agriculture fitright in with that.
But I also loved the, the nerdstuff with the computer science.
(02:17):
And I talked to Darren and hewas willing to give me an
internship for that summer wheremy main focus was working on
the first version of Adapt.
And from there, fast forward tothe next summer I worked with
GK again as an intern, alsoworking on Adapt.
However, it was a newer versionof Adapt where they completely
(02:37):
rebranded everything and then,following that, they decided to
give me a full-time offer inDecember.
I accepted it and started towork here in January.
Sarah (02:48):
So wow, there's a lot there, hudson, that's a lot.
So my first question for you ishow old were you when you went
to your first GK retreat, the GKTechnology Inc.
We go on a company retreatevery summer.
It's about a week long, it'sour company meetings for the
(03:09):
year, annual meetings, where wereally get a chance to like hash
through everything for the yearand where we're going to go for
the future.
And you know, as Hudson said,that's when family gets to come
in.
But how old were you on yourfirst company retreat?
Hudson Fuller (03:23):
I was 17.
Yep, some of them have beenthere a lot longer than I have.
It just never worked out thefirst couple of years of the
retreat for me to get to go.
But I got to go 17, right aftergraduating high school.
Sarah (03:37):
That's awesome.
Okay, and what was yourfavorite thing that you got to
do for fun on that retreat?
Hudson Fuller (03:43):
It was probably.
It was in Montana, up in theBillings area and we went to a
natural land bridge and myfavorite.
That was my favoritedestination that we went and did
and it was cool because I wasable to throw our rock across
the land bridge, get it acrossthe whole canyon.
Jodi (04:01):
I like that reasoning.
Sarah (04:03):
That's probably because you might know a little
something about baseball andthrowing a baseball.
Yep, I like that reasoning.
That's probably because youmight know a little something
about baseball and throwing abaseball.
Hudson Fuller (04:08):
Yep, I played college baseball for three years
and played throughout highschool and everything else at
Dakota State University.
Sarah (04:15):
Awesome.
So, speaking of Dakota StateUniversity, what did you major
in?
I don't think my computerscience description was really
encompassing of what your majorwas.
What did you major in, and wasthat first what you started in?
Tell us more.
Hudson Fuller (04:30):
Yeah, so my major was computer science actually.
Yep, you got it right and thenno, that's what I started with.
At first I was pretty sure Iliked computers enough to do it,
but once I got into some of theclasses where you actually were
hands-on doing stuff withcomputers, I realized that I
actually enjoyed doing it a lot.
(04:50):
It didn't feel like I was inschool, it just felt like I was
learning something for fun.
Sarah (04:54):
We are certainly glad that we have you around here to
solve all of our programmingissues, and you think it's fun.
That's great.
Jodi (05:03):
So what does that encompass?
Do you learn a little bit aboutcoding, about the hardware
Maybe?
What was your favorite classgoing through the whole computer
science degree program?
What was your favorite specificclass out of that?
Hudson Fuller (05:17):
Yeah.
So going through it was prettymuch all about coding and then
the theory behind coding.
It was pretty much all aboutcoding and then the theory
behind coding.
Some classes were every weekyou'd have some sort of coding
assignment where you'd have tomake something.
Other classes it was more aboutdo you understand how this idea
works, writing it out on apiece of paper.
(05:38):
And then, as far as my favoriteclass, it was probably in my
last semester, it was a softwareengineering class because it
was all about one big groupproject, basically the whole
semester, and it was my firstreal introduction to group work
in a code system.
But my two internships with GKI was kind of off on my little
(06:01):
lonesome working on the adaptbecause no one else understood C
sharp, which is the programminglanguage that adapt needs.
So it was awesome to be able towork through different pieces
at once to come together to afinal.
So it was an AI baseball pitchpredictor.
(06:32):
So like the purpose of it was topredict which pitch you should
use to trick the batter, or soit was more about say you knew
you were going to throw a sliderand then you could figure out,
based on how hard it was whereit was at in the strike zone and
what the count was.
(06:53):
It would figure out what themost likely outcome of that
pitch was, whether it wasgetting put in play a strike or
a ball.
Sarah (07:01):
So you could predict the strikes or the balls based on
what pitch you were going tothrow.
Hudson Fuller (07:06):
Yes, okay.
Jodi (07:08):
How accurate was it at predicting at the end?
Hudson Fuller (07:12):
It was between those three outcomes, it was, I
think, 80%, which is quite goodfor baseball, Because baseball
there's a lot of things that gointo baseball and lucky things
will happen that you can neverpredict.
Jodi (07:28):
That's really interesting.
So you kind of need a databaseto do that right.
Like you're kind of buildinglike a regression model and like
predicting response.
Where did you, where'd you pulldata from and like, were there
any factors that like came outthat were really predictive,
that you didn't think they'd be?
Or were you kind of limited on,like what you were pulling in
for?
Hudson Fuller (07:48):
Yeah, so there is a website called Baseball
Reference and it is for thenerds out there.
It has everything from how mucha pitch moves, exactly how hard
it is, to how many times perminute it's spinning, and there
is a library in Python thatallows you to pull all of this
(08:09):
data just almost instantly.
And you can pull back to, Ithink, it's 2012.
You could pull every pitch fromthat season, from a certain
team, from a certain player.
There's so many different waysthat you could pull in the data,
and that's where we struggledat first, because we could
export it in an Excelspreadsheet just to see it,
(08:30):
instead of trying to figure outby coding it what is actually in
there.
We thought that how much apitch moved was going to have
way more of an impact than itdid.
It ended up having almost azero impact.
It was way more about where inthe strike zone that pitch
actually was.
Sarah (08:49):
That's so cool, and did I hear you correctly a little bit
earlier that you just calledyourself a nerd?
Hudson Fuller (08:56):
Yes.
Sarah (08:57):
That's awesome.
I mean, we at GK actually liketo classify ourselves as geeks,
but that's okay.
I have also been known to beaccused of being a nerd as well,
so we're all in good companyhere.
That's awesome.
Hudson Fuller (09:12):
Nerds make the world go around.
Sarah (09:15):
So do geeks.
And here we are.
Jodi (09:19):
I love it because it's very similar to putting together
a science project.
Right you have a hypothesis ofwhat you think is going to
affect the outcome, right.
And then you test it.
In this case, you brought thedata back, you looked at it, you
built a model and found outthat it was actually a different
factor that influences it themost.
Sarah (09:35):
But on top of that is where you were talking about how
there's a lot of data that'sout there and, you know,
(09:57):
sometimes in baseball there'ssome things that are just
unpredictable and you just can'tpredict it.
That really isn't thatdifferent from agriculture.
There's a lot of things that weknow, there's a lot of things
that we deal with every day inthe field and we base a lot of
(10:21):
our decisions based in previousknowledge, in data that already
exists.
But sometimes there's thingsthat just happen, right yeah.
Jodi (10:26):
That's a great point, Sarah, Because like, okay, so
you've got all this data right,and I think we've been talking
in agriculture for the past oh Idon't know how many years ever
since we've gotten yieldmonitors on combines.
But we have a lot of data inagriculture now and it's kind of
piecing through and figuringout what actually has value and
what doesn't.
And that's kind of excitingwith AI right, because instead
(10:53):
of having to choose what factorsor like what data you're
pulling in to build your models,you can now kind of bring
everything in, as long as youhave a library that's organized
and I think we'll talk aboutthat with Adapt but you can kind
of look at all of it and figureout how that might predict
final yield at the end of theseason, and maybe that's a
little bit too optimistic atthis point in time, but that's
cool.
I think that's where the futureis is using artificial
(11:13):
intelligence to analyze a lot ofthe stuff that we have.
Sarah (11:15):
So lots of talk about baseball and data and how we're
doing predicting what's going onwith baseball.
So, but you've been reallyactive with baseball, so what
tell us a little bit about, likewhat exactly your history with
baseball has been?
Hudson Fuller (11:29):
Yeah.
So the first memory any of myfamily and me have with baseball
is me wanting to play t-ballwhen I was three, and we
couldn't until I was four, and Iwas so upset at that point in
time.
It was one of my sister'sfriends that was coaching that
team and it was hard for themnot to let me play.
But I had to wait one moresummer I did, and then worked my
(11:53):
way up from t-ball to coachpitch and then the first year
where I ever pitched at allwould have been probably when I
was 12, and I only think I threwtwice and it was something the
previous coach wanted me topitch more, but I didn't want to
, and then it just never reallycame up much until I.
It got mentioned once that I'dlike to try it, so we tried it.
(12:17):
It was okay, but it wasn'tfantastic.
Then we get up to the big fieldthe size that every team in
America plays once you're about14.
And then it was.
I got thrown on the moundagainst a bunch of 18 year olds
when I think I was 13.
And my second time ever doingit I had a line drive come back
(12:37):
and hit me in the wrist and itwas that we went to the
emergency room because wefigured it was going to be
broken.
It hit me that hard, Jeez Yep.
Then kept working through highschool and then I think it was
about my junior year I realizedthere was a chance I could play
college baseball and I wasreally interested in doing it.
So I emailed the coaches atDakota State because I knew
(13:00):
there was a chance that I wantedto do computer science and the
DSU has the best computerscience program in the area.
So I was like it would makeperfect sense if I would be able
to go play baseball there.
I got an email back sayingwe'll talk once you actually
turn into a senior, Got an emailback and went down for a visit
(13:21):
and got my scholarship offerthere.
Sarah (13:24):
Oh, that's awesome.
So you got through collegeplaying baseball.
Yes, that's really cool, that'spretty neat, that's that's
pretty cool.
So, and are you still?
Is baseball still a part ofyour life today?
Hudson Fuller (13:40):
Yes, so I actually coached the high school
team and the 16 year old teamhere in Clark.
It's awesome to be able to dothat and give back to the.
I've played with some of thekids too, because we've had to
play 13 year olds with the 18year olds, and then I also play
for our local amateur team inthe summers okay, so that, so
(14:01):
baseball still still a big partof your life and everything else
so it was.
Sarah (14:07):
It was always kind of interesting because, uh, I know
your parents followed you arounddoing taller games and
everything.
And it was always kind of funbecause I think Paul, your dad,
is Paul and he used to take offand he'd bring his computer with
and he'd map, you know, on theroad going all the way to see
you and everything like that andget a chance to catch up on the
(14:29):
baseball and then come on backagain.
So it's, it's interesting.
We've been hearing about Hudsonand his baseball as as
employees of of GK for for alittle bit.
Well, that's pretty neat.
I think Jody was asking acouple of questions about ADAPT.
So your first internship withGK was ADAPT and your second
(14:51):
internship was ADAPT, but alldifferent ADAPT, correct?
Hudson Fuller (14:55):
Yes.
Sarah (14:56):
Why don't you tell us a little bit about that?
What is ADAPT?
What does it do?
Yep.
Hudson Fuller (15:00):
Everything else.
So ADAPT is a standard thatAgGateway has set in an attempt
to unionize data formats acrossall of the agricultural software
companies.
So everyone can say, say, putit into a John Deere monitor and
pull it back out, and they'llbe able to turn it back into the
data they want to use in theirown software system.
(15:23):
The goal of it has always beenfull unionization and a lot of
different small companies havesigned on to make it a
possibility.
There are not that many thatactually have ADAPT working at
all and it's because it keepschanging, which is a great thing
, because it needs to keepupdating.
(15:46):
So the first version of ADAPTwe'll call the ADAPT framework.
That is the one that I think itwas.
Three or four years ago theidea was brought on to make a
standardization and, darren, heknew about it, he wanted to
implement it.
He just did not understand thecoding language that it ran with
and then it happened to workout perfectly that I did for my
(16:09):
first internship.
So we worked through that firstsummer and with that, with the
ADAPT framework, it could exportinto multiple different file
types, that being GeoJSONs, iso,xmls and then also an ADAPT
file, itself known as an ADF.
Sarah (16:27):
Okay.
So when you're getting andthat's the as applied data, then
pretty much that you're talkingabout coming from, okay, from
controllers, getting into astandardized format that we can
just bring in as generic fileformats.
Hudson Fuller (16:41):
Yes, yeah.
And then from there we got thatsomewhat working at the end of
the summer.
However, there was a few issueswith it with it.
And then we realized that adapthad changed something in their
code that made it not securelysigned, meaning we could not
release it to the public.
And then come to the nextsummer we know we figure out how
(17:06):
to fix that, I get that working.
And then we realized that adaptchanged everything.
They decided to get rid of thedifferent types of files that
Adapt would export to, and justalways export to, a JSON file,
which is it's not a bad idea atall, because then you don't have
(17:26):
a whole bunch of different filetypes floating around.
But that meant I had to recodeeverything so we could export
into that JSON style.
Jodi (17:36):
So, backing up even before Adapt, can you tell us a little
bit more about what the problemwas and why is there this drive
to have something standard?
And I think in my head I knowsometimes we get asked as
employees at GK to handledifferent information types and
I know sometimes I just can'topen them right away or like I
don't have the right import.
(17:56):
Can you tell us more about,like, what the problem at hand
is here and then how ADAPT andthat framework is helping to get
us closer to being able to viewand use this data?
Hudson Fuller (18:10):
Yeah.
So it's as simple as instead ofhaving a whole bunch of
different fruits and trying tomake a big fruit salad, you just
have one type of fruit set outat a buffet.
There is no real point in anyindustry, but in the agriculture
industry it's been waydifferent, because there are
more file types.
There's the shapes and everysingle different file type
(18:32):
associated with the shapes.
There's the shapes and everysingle different file type
associated with the shapes.
Every single one of those justleads to more space taken up on
a computer and leads to morethings having to get put on a
flash drive or whatever else toget put into a monitor.
If everything could get put inas one, our space would be a lot
(18:57):
smaller and there would be noconfusion within the industry.
Everything, everything fallsunder just a few file types,
instead of 50 or 100 that mightbe out there even though we
don't use them.
Sarah (19:02):
There are so many out there that are not used in our
area of the world but might bein australia, asia or anywhere
else so kind of like hydraulichookups back in the day then,
like John Deere had their ownspecialized hydraulic hookups
for many, many, many years andyou couldn't hook up like if it
(19:25):
was a chisel plow or somethingfrom a different brand.
You couldn't hook it up to theJohn Deere unless you had the
converter to convert thehydraulic hookup.
Hudson Fuller (19:36):
Yes, that's a perfect analogy.
Sarah (19:39):
Once we standardized hydraulic hookups, now we could
hook up whatever equipment wewanted to whatever tractor we
wanted, without converters,because we all had the same
hookups.
Hudson Fuller (19:51):
Yes.
Sarah (19:52):
Every color would talk to every color.
Hudson Fuller (19:54):
Yeah, that is a perfect analogy for that.
Jodi (19:58):
Would this get us to be where, like we wouldn't have to
use the FOD importer for yielddata that kind of thing anymore,
or I suppose, like there'dstill be older controllers that
would export like that correct?
Would this be something thatcomes out in newer controllers
and we'd see that newercontrollers are using the ADAPT
standard, or is this like astandalone?
How does ADAPT get implemented,I guess would be my question.
(20:21):
What do things look like onceADAPT is implemented?
Hudson Fuller (20:27):
Yeah, so the idea is to not truly replace FOD
because the older monitors andcontrollers they will never get
updated to ADAPT Okay.
But a lot of the newer ones Iknow specifically with Case they
have the full implementation ofADAPT within their newer
monitors.
I don't know the numbers thatcontroller monitors are, but I
(20:50):
know they use ADAPT and that'swhy we've been pushing hard to
get the ADAPT framework out,because they use the ISO XML
plugin of Adapt so that itexports into a task data into
that style format.
But it all uses a format insidebuilt from Adapt.
Sarah (21:09):
So would we be exporting out a geo JSON then essentially
the ADAPT system that the CASEcontroller could read.
Hudson Fuller (21:17):
CASE uses an ISO XML file that has the boundary
and it has the averages andcertain other numbers associated
with your prescriptions, andthen it also exports a bin file,
and that bin file has a gridassociated with all of your
(21:41):
different prescriptions forvariable rate for that field.
Jodi (21:44):
That's really interesting.
I guess I'm saying it'sinteresting because I've never
thought about you know, becauseI think right now for exporting
to cases you know, or likeimporting yield data, it comes
back in like a CN1 file or likea.
My point is like, even with likeKMLs too, when anything is
(22:04):
different from a shapefile, Idon't know how it works right,
like the thought of like havinga bin that has the grid of like
all of the prescriptions, itjust kind of blowing my mind.
Right now I'm just trying tothink through, like what are
could you talk about like someother types of like file types
and like how that's set up?
(22:25):
Like could you walk through,like in your computer language,
computer science knowledge, canyou talk about like what is a
shape file?
And then like what does a caseprescription file look like?
And maybe like another exporttype that you can think of?
Because I think when I knowwhat these things kind of act
like, it helps me reallyunderstand, like how to export
(22:47):
things better out of ad master,like why I need to jump through
a certain hoop to get what Ineed to get done yeah, so we'll
start off with a shapefile andwhen I'm actually working with
the data from a shapefile Idon't look at the shapefile.
Hudson Fuller (23:00):
I'm not seeing the boundary or the polygon
style prescription.
I'm not seeing that at all.
I am just seeing the data inthe back end.
So we have a tool within ADMSthat allows us to read in all of
those points on the shapefile.
Usually it is in differentfeatures.
(23:21):
That is like in the backendwhat a shapefile has.
Each polygon is not a polygon,it is a feature and each of
those features has so manydifferent values associated with
it.
If it's relating to aprescription, they will have
some sort of numerical valuevalue say it's 112 pounds per
acre, whatever it may be andthen they'll also have an extent
(23:45):
, is what we call it, and thatis the edge of that feature.
So every little square has fournumbers and that is the lat
long of each corner of that.
And then in ADMS we always readit in as northern and easting
in the UTM style.
Sometimes we have tore-transform it back into lat
(24:09):
long to make it so that we canuse that data across the world,
not needing to know the UTM zone.
That specifically has come uprecently when I've been working
with the field info stuff andtrying to figure out how can we
best actually get where thisfield is.
But that's a different topicfor a different time.
(24:30):
And then the Adapt ISO XML files.
When you open up that file, theISO file, you don't actually
see a picture of the boundary.
You will see probably a hundredpoints or so that make up the
edge of that field's boundary.
Along with that you will havethe average of the prescription,
(24:54):
like what the average poundsper acre in that field is.
You will have a default valueand a out-of-field value.
These values are in there.
So if for some reason thecontroller cannot use the bin
file, which is a file that justyou can't actually see the data
in there, the computer can seethe data though it knows exactly
(25:17):
what each point in that fieldthe prescription can be.
And then really another thirdtype that gets used is a GeoJSON
, which we also, through ADAPT,will have the ability to export
a prescription from.
That comes out as two separateGeoJSON files in one folder.
(25:38):
The one contains the basicinformation, such as the
boundary, the default rates, theout-of-field rates, and the
other, geojson, will be a very,very long JSON file that has a
whole bunch of differentboundary points in it or
prescription points, and then itwill have the variable rate for
(25:58):
each point within that field.
And, for example, one of thetest fields that we have through
GK, the Bloom House one, thatone ends up at 25,000 different
points of information that endup getting thrown into that
GeoJSON file.
Sarah (26:15):
Wow.
Hudson Fuller (26:16):
But the controllers that read that can
do it in an instant.
Jodi (26:40):
Wow that the controllers that read that can do it in an
instant.
When you increase theresolution or like decrease it
down, where you're going fromlike three meters to two meters
or like 10 meters to sevenmeters, you're increasing the
number of data points that needto be collected for each of
those, the squares, and Isuppose, for like older monitors
, like we see in, like themanual for exporting, um, that
that matrix, right, some of theolder controllers can't handle
that much data, right, right,and we've got to reduce that
(27:01):
polygon size.
But that makes sense If we'vegot data at all those points,
that takes up more space andmore memory with the computer to
read those and process whatthat actually means.
To tell the spreader or theplanter in the back what to do
and how to vary the rate, thespreader, the planter in the
back, what to do and how to varythe rate.
Sarah (27:21):
Gosh, I'm so glad that you're talking with us about
this because you know, from ourperspective, or like for me,
especially as agronomists, I seea lot of these things on the
front end but I've got no ideahow it's actually working in the
computer to see these thingsand convert it into something
(27:41):
that is, you know making ratesvary in the back, so this is
really fascinating for me tolearn from you, hudson, and not
only making the rates vary, butmaking the rates vary in the way
that makes you know theagronomist has the rates set up,
but we want to keep theintegrity of that prescription,
and not every single exportprocess can do that well, and so
(28:04):
it's really exciting to hearabout how this all comes
together, and not only that, butto have the explanation of all
of those different file types.
That's pretty fun.
Jodi (28:15):
So we're going to cut it here, but we've had such a
fantastic conversation justgetting to know Hudson a little
bit better, learning about hisbackground and how baseball has
helped make Hudson a bettercomputer programmer.
But we've also talked a littlebit about the ADAPT standard and
some of the things thatHudson's working on, and it's
been such a pleasure to haveHudson explain these things to
(28:36):
us and I know that we're goingto do the same when we come back
with Hudson in the next episode, in the second part of our talk
with him.
So thank you all so much forjoining us today and we will see
you on the next full-lengthepisode of Ag Geek.
Speak with GK Technology.
Sarah (28:53):
And at GK Technology we have a map and an app for that.
Popular Podcasts
United States of Kennedy
United States of Kennedy is a podcast about our cultural fascination with the Kennedy dynasty. Every week, hosts Lyra Smith and George Civeris go into one aspect of the Kennedy story.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.