One Designer + AI. Hundreds of Deliverables. — Vincent Wendy, AI Engineer
Read the talk
One Designer, Hundreds of Deliverables: Building a Conference Design Workflow with AI
Vincent Wendy explains how explicit design rules, reusable graphics, data-driven exports, and visual checks help one designer support a large conference—and why handling exceptions remains central to the job.
From a talk by Vincent Wendy
At a glance
Ideas worth remembering
Explicit typography, colors, and spacing make reusable design possible and give an agent concrete constraints for implementing it.
Reusable tools turn repeated work into selection, data loading, and export. Room signage and speaker graphics illustrate that approach, while the flash-drive delivery shows that some operational steps remain manual.
Visual matching can reduce photo-search effort and add a check for missing sponsor logos. The reported accuracy reflects the speaker’s experience and tests; the talk supplies no general error rate.
Automation must accommodate exceptions. The missing edit button demonstrates why a workflow needs a practical path from a last-minute change to a corrected deliverable.
The scale problem is also an error problem
The talk begins as a practical account of designing AI Engineer conferences for an audience made up mostly of engineers. The senior creative designer works in a team of roughly 12 to 15 people, with one designer responsible for a wide range of conference materials. Automation is his answer to the mismatch between that small team and the event’s growing demands.
When he prepared the presentation, the conference expected 6,000 attendees; by the talk, that number had reached 7,000. It also had more than 140 sponsors, more than 300 speakers, and more than 600 sessions. Those counts generate different kinds of work: sponsor assets, speaker graphics, stickers, landing pages, track mascots, physical signs, and digital signage. Scaling design therefore means supporting many formats and many changing records.
The risk grows with the detail count. A missing sponsor logo creates a serious omission, while an incorrect speaker schedule can misdirect people. His observation that a thousand details create a thousand ways to fail establishes the real requirement: a workflow must produce enough assets while preserving completeness and correctness.
Suggest correction
This note stays in this page until you copy or download it. Nothing is submitted; reloading clears the draft.
Combine tools around the deliverable
He introduces his design team as himself, Devin, GPT, and Figma. His working premise is that available tools have become capable enough for concrete problems to drive the work. Asked what inspires his conference design, he points to the problems themselves: the operational need gives him something specific to solve.
A pelican riding a bicycle illustrates how he works around a capability limit. He describes testing a model’s ability to create that scene directly as a vector file and finding the basic model’s output unusable for his design work. His alternative is to ask ChatGPT for a still image, such as a PNG, and then vectorize it in Figma. The intermediate format changes, but the workflow can still reach a shippable asset. He does not provide a comparison of vector quality or editing effort; the example teaches a way to compose tools rather than a guarantee about conversion fidelity.
Suggest correction
This note stays in this page until you copy or download it. Nothing is submitted; reloading clears the draft.
Define the design system before generating variations
His method has five parts: establish the foundation, make designs reusable, automate workflows, validate output, and remove friction. The foundation consists of the design system’s typography, colors, and components. The website embodies those decisions, including primary and accent colors, typography, and the tagline, giving subsequent work an established set of choices.
He connects this approach to his product-design background and atomic design: create the smallest useful parts, then combine them like LEGO pieces into complete deliverables. The mechanism is reuse through composition. A designer establishes the pieces and their relationships once, making it possible to assemble further outputs without deciding every visual detail again.
Those rules also constrain model output. He explicitly supplies desktop and mobile typography because, in his experience, language models otherwise introduce arbitrary font sizes. Defining the values reduces the room for inconsistent choices. The foundation thus serves both as a shared visual language and as a specification the model can follow.
Suggest correction
This note stays in this page until you copy or download it. Nothing is submitted; reloading clears the draft.
Reuse the brand and render schedules from current data
Once the website expresses the branding, other teams can use it as the basis for their own materials. He describes marketing creating email designs, flyers, and documents from that reference. Repeated mascot designs follow the same principle: an established visual treatment can be applied again. This distributes production work while retaining a common foundation.
Room schedules provide the first automation example. Previously, he laid them out manually in Figma. The newer workflow uses Devin to pull the latest data for a requested room and set of days, then exports the result as a downloadable PNG. He transfers the image to a flash drive and puts it on the room’s screen. The described pipeline automates data retrieval and graphic production, while display delivery still involves a physical transfer.
He reports that the resulting schedule data is accurate, but the talk does not explain the data source or how accuracy is checked. Pulling current records addresses the burden of manually rebuilding schedules; it does not, by itself, establish that the underlying records are correct.
For visual fidelity, he describes shortening the feedback loop between design and implementation. Earlier handoffs to engineers could require repeated corrections when the result did not match the design precisely. With Devin, he can ask for a more accurate implementation, connect it to MCP, or provide a specification sheet defining spacing and font sizes. The concrete measurements give the agent something more precise to follow than a general request to match the design.
Suggest correction
This note stays in this page until you copy or download it. Nothing is submitted; reloading clears the draft.
Turn speaker graphics into a reusable product
With more than 300 speakers, producing announcement graphics individually would overwhelm his capacity. He instead demonstrates a speaker-announcement tool with a selection control, editable name text, and a landscape mode. When a speaker has a headshot and the necessary details, he says the tool can export automatically. Trading cards extend the same approach to another deliverable, and he identifies TBPN as the inspiration for one example.
He calls the outputs pixel perfect. His explanation centers on how design information reaches the implementation: Devin lives in Slack, so his working loop moves between Slack, Figma, and Slack again before shipping. He contrasts that direct loop with the more extended research, design, and feedback process he used earlier in his product-design career. He presents this as a change that suits his own work, rather than demonstrating that research is unnecessary for every product.
The specification can arrive through MCP or through a document. He describes a free Figma plugin that adds annotations to a PDF, exposing spacing, font sizes, and colors. Even when layers have generic names such as frame three or frame four, he says the model can interpret the specification. The useful mechanism is the explicit visual measurement; the talk gives no measured tolerance for the claimed pixel-perfect result.
Suggest correction
This note stays in this page until you copy or download it. Nothing is submitted; reloading clears the draft.
Find the right speaker photograph
A newer task is finding event photographs for speaker thumbnails. He asks Devin who appears in a photo and describes an interface for deciding whether the pictured person matches a speaker. In the demonstration, he confirms a match as Jason Liu. Previously, thumbnail work involved searching through the photographer’s image codes one by one, sometimes using the time as a clue. A proposed identity now leads him directly to a photo he can download and place in the thumbnail.
He considers the matching fairly accurate, but provides no error rate, evaluation set, or explanation of the recognition method. The demonstrated value is a reduction in search effort with a person checking the proposed match. That experience reinforces his broader enthusiasm: fast design and automation make a worthwhile problem the starting point for useful work. His suggestion that almost anything can then be solved expresses that optimism, rather than establishing a technical capability boundary.
Suggest correction
This note stays in this page until you copy or download it. Nothing is submitted; reloading clears the draft.
Use visual comparison to catch missing logos
The sponsor banner in the lobby contains logos for more than 140 sponsors. He asks Devin to compare the graphic and check for missing logos, then reports 100% accuracy in the tests he ran. He also uses the same technique on the conference T-shirt. This makes visual inspection part of the production workflow: the agent checks whether a dense graphic contains the required sponsor marks.
The accuracy claim is limited to his own tests. He does not give their number, the comparison reference, or the conditions under which omissions were tested. His practical recommendation is to combine human and AI inspection: a person can overlook a small missing element, and an additional visual check can catch it. The agent supplies another opportunity to detect an error before the design reaches its audience.
Suggest correction
This note stays in this page until you copy or download it. Nothing is submitted; reloading clears the draft.
Design the attendee journey and make exceptions manageable
Removing friction begins with imagining the experience as a user. He mentally walks through the attendee’s path to registration, wayfinding, QR codes, and rooms. These elements need to connect so people know where to go. The design requirement extends beyond the appearance of individual signs to whether the sequence of information supports navigation.
He then identifies handling exceptions as the real job. The schedule tool initially lacked an edit button. One morning, a schedule needed updating, so he asked Devin to add the control. He reports that it did, allowing him to change the schedule, export a revised PNG, and put it back on the screen. The example shows both a gap in the first tool and the usefulness of being able to extend it when a concrete need appears.
The update still requires exporting and replacing the displayed image. The talk also leaves open how those edits relate to the original schedule data. What he establishes is a shorter path from an unexpected operational requirement to revised signage, which he finds much easier than the earlier manual process.
His closing advice is to solve scale by thinking small: identify the smallest things that can go wrong and try to address them in advance. He ends by returning to the value of having a real problem, because it gives the work a direction and a reason to ship a better product. His broad claim that everything can be automated remains an expression of confidence; the actionable lesson is to focus on specific failure points and make them easier to handle.
Suggest correction
This note stays in this page until you copy or download it. Nothing is submitted; reloading clears the draft.
Read the complete timestamped transcript
- 0:12
All right.
- 0:14
Hello everyone. Hope you guys having a
- 0:16
good time at the conference.
- 0:18
So, before we start
- 0:20
how many of you are actually uh
- 0:22
designers? Like a product designer. Hey,
- 0:24
one hands and another. Okay.
- 0:28
And how many I assume that the rest of
- 0:31
you are engineers? Is that correct?
- 0:33
Yeah, pretty much. Okay.
- 0:35
So, today's talk is a non-technical
- 0:38
talk, but more of a real-world
- 0:40
experience how I created the design for
- 0:44
AI Engineer this conference and
- 0:46
our other past conference as well and
- 0:48
how AI has helped me. And so, the talk
- 0:53
today is one designer plus AI, which is
- 0:56
me as the designer,
- 0:57
and hundreds of deliverables.
- 1:00
All right, let's start.
- 1:02
So, my name is Vinson Weng. I am a
- 1:04
senior creative designer at AI Engineer.
- 1:07
And at AI Engineer, it's a very small
- 1:10
team. So, we only have around 12 people
- 1:14
to 15 people at the moment. And
- 1:17
everyone has been doing their own thing
- 1:20
and
- 1:21
I think AI has been like has been a
- 1:24
really helpful way to like helping
- 1:27
everybody doing everything.
- 1:29
And
- 1:30
at if for if an at this scale
- 1:33
we have a problem, obviously, right?
- 1:35
And the problem is the scale problem or
- 1:38
I would call the challenges.
- 1:40
And how to overcome it?
- 1:43
It's basically automation and we get to
- 1:46
that in the later part of this talk.
- 1:49
So,
- 1:52
when I prepared this talk, we only
- 1:54
expected 6,000 attendees and now it's
- 1:56
7,000. Well,
- 1:58
good for us.
- 2:00
And then we have 140 sponsors. More.
- 2:04
140 plus sponsors. And then 300 plus
- 2:07
speakers, 600 plus sessions, and one
- 2:09
designer.
- 2:11
And everybody needs
- 2:14
every designs, right? Like every single
- 2:16
thing needs design. Sponsor needs
- 2:18
assets, speaker needs graphic.
- 2:20
You need sign it so you don't get lost.
- 2:23
And
- 2:25
this is basically what we do, what I do.
- 2:28
So, from stickers, do you like your
- 2:30
swag, your stickers?
- 2:32
Well, I hope you do because I create
- 2:34
that design, too. And
- 2:37
to a landing page,
- 2:38
speaker announcement, track mascot, all
- 2:41
the stuff that you see,
- 2:43
most of the stuff that you see
- 2:45
here, from a sign it to a
- 2:48
digital sign it, landing page,
- 2:50
everything is a deliverable.
- 2:52
And
- 2:55
a thousand details means a thousand way
- 2:57
to fail, right?
- 2:58
Because
- 3:01
I'm missing sponsor logos, going to be a
- 3:03
huge issue. And speakers that have a
- 3:06
wrong schedule, also a huge issues,
- 3:08
right? And it seems impossible to handle
- 3:12
that many kind of deliverables, but
- 3:15
yeah, meet my design team.
- 3:17
So, it's me and Devin, GPT, and Figma.
- 3:26
And right now we are at the stage where
- 3:28
tools isn't the like it's not a problem
- 3:31
anymore, but having a real problem is
- 3:33
our advantage.
- 3:34
So, for example,
- 3:37
when someone asked me, "What inspired
- 3:39
you when designing in AI engineer?"
- 3:41
I don't know the answer back then, but
- 3:43
after I think about it, it's actually a
- 3:45
problem that inspired me to like
- 3:47
designing in this AI engineer. And we'll
- 3:50
get to that in the latter part of this
- 3:53
talk.
- 3:54
So,
- 3:56
have you guys seen the talk by Simon
- 3:58
Wilson like in 2025?
- 4:00
>> Yeah.
- 4:02
>> Yeah, and it's pretty interesting,
- 4:03
right? He asked to
- 4:06
He asked every LLM to create
- 4:09
an a vector file, which is basically a
- 4:12
pelican riding a bicycle.
- 4:14
And it is basically to test and I tested
- 4:16
again and it's still doing this for the
- 4:19
basic model.
- 4:21
And it's not usable for me as a
- 4:22
designer. But as a designer, we have to
- 4:25
think outside the box.
- 4:27
And we could simply ask ChatGPT create a
- 4:30
still image like a PNG for a pelican
- 4:32
riding a bicycle and then I can
- 4:33
vectorize it on on Figma.
- 4:36
And we can ship that now.
- 4:38
So, we have to think outside the box
- 4:40
here
- 4:41
regardless the capabilities of the LLM.
- 4:44
And
- 4:46
So, how to solve this scale problem,
- 4:49
right?
- 4:52
Basically five five things. So,
- 4:54
foundation first, reusable designs,
- 4:57
automated workflows,
- 4:58
validated output, and also remove
- 5:00
frictions.
- 5:02
The foundation is definitely the core
- 5:05
part that we need to set up right. Like
- 5:07
the design system, typography, colors,
- 5:09
components, like other stuff.
- 5:12
And once this is set up, like for
- 5:15
example, when we create the website,
- 5:17
it's all set up within this thing.
- 5:20
And yeah, this is just an example. Like
- 5:22
we have the colors, primary, and then
- 5:25
also the accent colors,
- 5:27
the typography,
- 5:29
and also the tagline, all the other
- 5:30
stuff.
- 5:32
And
- 5:34
also, have you guys Are you guys
- 5:35
familiar with the atomic designs?
- 5:38
So, yeah, my previous background is I'm
- 5:40
a product designer. So, I'm pretty
- 5:42
familiar with the
- 5:44
thing where we need to create a
- 5:46
user-centric design and also like atomic
- 5:48
designs, right? Where we create the
- 5:49
smallest part possible and then
- 5:51
combining it into like basically a LEGO
- 5:53
pieces and then into a deliverables.
- 5:57
And this is pretty useful in my job desk
- 6:01
right now.
- 6:02
So, once we set up all of those
- 6:06
foundation, we basically need to create
- 6:09
for example, we use Defont a lot. In at
- 6:11
the office, we everybody use Defont.
- 6:14
Everybody like
- 6:16
abusing Defont for example.
- 6:17
Yeah. And
- 6:20
So, in this case, I just need hey, we
- 6:23
use this desktop typography and this
- 6:25
mobile typography because we know
- 6:29
Cloud or like any other LLMs love to
- 6:33
like throwing some random
- 6:35
font size, right? And if we don't define
- 6:37
it, it just delivering a slope like the
- 6:40
previous slope.
- 6:42
And
- 6:44
yeah.
- 6:45
Typography, color and stuff and
- 6:48
and then it comes to reusable design.
- 6:51
So, once we set up it right, like the
- 6:53
website is has the
- 6:56
the branding to it, all the other teams
- 6:59
on the AI engineer, like for example,
- 7:01
the marketing teams
- 7:03
can create everything basically. Like
- 7:05
they can create an email design based on
- 7:07
that. They can create a flyer, a
- 7:10
document
- 7:11
just based on the website because it's
- 7:13
already defined
- 7:15
like it defined early.
- 7:19
And yeah, once you get the design, you
- 7:21
can just rinse and repeat.
- 7:24
For example, the mascot, it's all has
- 7:26
the pretty much the same design and it's
- 7:28
rinse and repeat. And if you
- 7:33
already defining those things, you can
- 7:34
basically like create one design that
- 7:36
works for all.
- 7:38
And this is the part that I'm most
- 7:39
interesting to talk about, which is the
- 7:42
automated workflows.
- 7:44
Before, for example,
- 7:46
if you take a look outside the room,
- 7:48
there's a schedule, right? The schedule
- 7:50
for each and everyone.
- 7:52
So, we used to do it manually on Figma,
- 7:56
but now we use Devin for it.
- 7:58
And
- 8:00
let me show you.
- 8:04
Hey.
- 8:06
So, right now we just pull the latest
- 8:08
data. I just asked Devin like, "Hey,
- 8:11
I want this room
- 8:13
at these days." And then we can just
- 8:16
export it, download it PNG, and the data
- 8:18
is all accurate, and then we can just
- 8:21
ship it to the flash drive, and then
- 8:24
put it on the screen.
- 8:25
And
- 8:27
it was like impossible before because
- 8:29
the friction is just too much between
- 8:32
the designers and the developers.
- 8:34
We [snorts] cannot make things like
- 8:35
pixel perfect because
- 8:37
once we tell the designer, "Hey, this is
- 8:39
the design." And then
- 8:41
Sorry, the the engineers that created
- 8:43
the design, for example. "Hey, I need
- 8:46
this to be delivered." And then they
- 8:48
don't create it pixel perfect, it's a
- 8:50
lot of
- 8:52
feedback loop, right? But with Devin, we
- 8:54
just say, "Hey,
- 8:57
can you make this more accurate?"
- 8:59
We can just connect it to MCP, and then
- 9:01
if it doesn't work, we can just always
- 9:04
like
- 9:05
give a spec sheet or something that
- 9:08
can be defined like what's the spacing,
- 9:11
what's the
- 9:12
font size, etc. And
- 9:17
this is what we do to for the speaker
- 9:18
announcement. So,
- 9:21
we have 300 plus speakers, and it's
- 9:23
impossible for me to like handle one by
- 9:25
one, right? So, we create this thing,
- 9:28
which is called which you can also
- 9:30
access to speaker announcement, and you
- 9:32
can also try
- 9:34
it yourself.
- 9:36
Like this one, for example.
- 9:38
You can select it right here.
- 9:40
And then you can also change your name.
- 9:42
Well, that
- 9:44
Yeah. For example, this you can change
- 9:46
the name to whatever you want. And we
- 9:49
also have the landscape mode
- 9:51
which can be also loaded. If the speaker
- 9:53
also have the headshot and all the
- 9:56
details, it will automatically export.
- 9:59
And we also have the trading cards
- 10:01
which is surprisingly pretty popular.
- 10:05
And we have a different team. And this
- 10:07
is all pixel perfect.
- 10:10
All right. For example, this one.
- 10:14
This is inspired by TBPN, so
- 10:19
Yeah. And how do I deliver this in pixel
- 10:22
perfect? Let's jump into it.
- 10:25
So,
- 10:27
the process here is
- 10:31
before
- 10:32
when I start my career as a product
- 10:35
designer, it used to be just
- 10:37
okay, we need to research, we need to
- 10:40
build product like design thinking in
- 10:41
general, right? And then feedback loop
- 10:44
and stuff like that. But right now, it's
- 10:47
it's just outdated for me. Like in my
- 10:50
case,
- 10:51
we just go to Slack,
- 10:53
Figma, and then send it back to Slack
- 10:55
because Devin or Devin live in Slack,
- 10:58
and then ship all the things that he
- 11:00
need.
- 11:01
>> [snorts]
- 11:01
>> Like for example, if we can connect the
- 11:04
MCP or also the
- 11:06
spec document, which is for example the
- 11:09
spec sheet like this,
- 11:10
which is
- 11:11
uh plugin in Figma if you interested.
- 11:15
It's free and it's basically give an
- 11:17
annotation to the PDF.
- 11:20
And
- 11:22
yeah, all designers don't name their
- 11:24
layers, so yeah, this is just like
- 11:27
some random frame three, frame four, but
- 11:29
the LLM will get it.
- 11:31
And it's basically defining all this
- 11:34
spacing, all this
- 11:36
font size, and then all the
- 11:39
colors and stuff. It's definitely going
- 11:40
to help you develop a pixel-perfect
- 11:43
product.
- 11:44
And
- 11:47
we also have just recently like today
- 11:49
have uh photos which we have to create
- 11:52
the thumbnail for its
- 11:54
speaker, right? And then we ask Devin
- 11:56
like, "Hey, who is this person?" And
- 12:00
yeah, it kind of did. Like I make a
- 12:03
Tinder kind of
- 12:05
you know, detection
- 12:07
if this is the same person or not. And I
- 12:09
think it's pretty accurate.
- 12:11
It's Jason Liu. Yes.
- 12:13
And then we can use this to like
- 12:17
for context. Like before, when we create
- 12:19
the thumbnail, we have to search all the
- 12:21
codes that photographer have and search
- 12:24
it one by one and maybe by time
- 12:26
if possible. But now we can just like,
- 12:29
"Oh, this is Jason Liu. Download that
- 12:31
photo." And then we can paste it into
- 12:32
the thumbnail, right? And it's pretty
- 12:35
amazing. I mean, the world that we live
- 12:38
in right now is actually like the state
- 12:41
for me as a designer is already at the
- 12:43
peak because
- 12:45
what what else can you ask for, right? I
- 12:48
mean, we already have things to
- 12:49
automate, we already have things to
- 12:51
create the design fast.
- 12:53
Basically, all you need is a problem
- 12:56
because once you have a problem that
- 12:57
worth solving, you can
- 13:00
basically solve anything.
- 13:02
And back to my talk, I got sidetracked
- 13:05
right there.
- 13:06
Yeah.
- 13:07
And then yeah. And this is also the
- 13:09
amazing thing that we test.
- 13:11
So, as you know, we have like hundreds
- 13:15
of sponsors, right? Like 140 plus. And
- 13:18
as you can see on the at the lobby, we
- 13:21
have the banner with all the sponsors.
- 13:24
And
- 13:25
I basically tell Devin like
- 13:27
"Hi, could you compare
- 13:29
could you check if there are any missing
- 13:31
logos in this graphic?" And the accuracy
- 13:34
is 100% based on the test that I
- 13:38
do. So,
- 13:40
which is pretty well. And we use the
- 13:41
same thing for the
- 13:45
T-shirt that you got for your swag.
- 13:48
And yeah, surprisingly, Devin knows how
- 13:52
to like visualize things, right? Like
- 13:54
how to detect things visually.
- 13:57
And that is very surprising because
- 13:59
as a human, we can like give errors. Oh,
- 14:03
turns out there's one small something
- 14:04
that is missing. But with this kind of
- 14:07
thing, we can like double-check. So,
- 14:09
human plus AI, combine it,
- 14:13
well, you got your own QA team.
- 14:15
And then remove fiction.
- 14:17
So, this is just uh the way of thinking.
- 14:21
So, as a designer, we have to think
- 14:24
as a user, not as a designer, all right?
- 14:27
Because every user has its needs.
- 14:29
You can walk through the for example,
- 14:31
the map plan here. So, basically,
- 14:35
I'm imagining myself as an attendee to
- 14:37
go to the registration, go to the
- 14:41
see the wayfinding and the QR code and
- 14:44
then all the stuff. Basically,
- 14:46
everything needs to be connected so you
- 14:48
guys don't get lost and knows how to
- 14:51
find your rooms and
- 14:54
other stuff.
- 14:55
And
- 14:57
the real job is handling exceptions.
- 14:59
So, for example, oh, I have Yeah.
- 15:03
>> [snorts]
- 15:03
>> For example,
- 15:04
um
- 15:06
there is a schedule update, all right?
- 15:09
And
- 15:10
when we create this thing, it doesn't
- 15:12
has an edit button. And then one
- 15:15
morning, it just "Hey,
- 15:17
this schedule needs to be updated and we
- 15:19
don't have those edit buttons." I could
- 15:21
just ask Devin, "Hey, can you add me an
- 15:23
edit button?" And then it did. So, we
- 15:26
can change everything now and then ship
- 15:28
it to PNG and replug it to the screen,
- 15:32
which is pretty convenient, right? And
- 15:34
those exceptions, right? It
- 15:37
it's not possible before
- 15:39
when we have to do it manually and
- 15:40
stuff.
- 15:41
But now it's just get easier. And
- 15:46
so, the takeaway here is that to solve
- 15:49
the scale problem, you have to actually
- 15:51
think small. Think all the smallest
- 15:52
thing possible. Think everything that
- 15:55
can go wrong and will go wrong and then
- 15:57
try to solve it before. And also, like
- 16:02
yeah, right now basically you can
- 16:03
automate everything.
- 16:05
And
- 16:08
at this moment, having a problem is
- 16:10
actually going to benefit you because
- 16:12
that's going to help you ship
- 16:14
a better product, going to ship uh
- 16:16
things that are
- 16:17
good. And yeah, I think that's all that
- 16:20
I can share. Hope my talk has some
- 16:23
benefits to you and
- 16:24
yeah.
- 16:25
That's all. Thanks, guys.
- 16:27
>> [applause]
- 16:46
[music]