July 9, 2026 • 17 mins
In this episode of "Full Tech Ahead," host Amanda Razani interviews Chris Yates, SVP of Product, Design and Engineering at Pantheon. They discuss the critical "trust factor" and governance bottlenecks emerging as organizations rapidly adopt AI tools.
Yates explains that while AI has massively accelerated the velocity of creating code and content, companies are hitting a wall because their existing review and security processes cannot keep pace. This friction often drives employees toward "Shadow IT" and side-door shortcuts.
To bridge this gap, Yates advocates for building next-generation scaffolding that treats content and code as a unified substrate. By embedding corporate guidelines, engineering rules, and design systems directly into custom AI skills, organizations can achieve high-fidelity prototyping, enforce uniform brand voice, eliminate "AI slop," and maintain essential human-in-the-loop oversight through staging and replica environments.
Key Quotes
- "Pantheon is, as we say, where the web works. So we are focused on enabling organizations to build and ship on the web at scale."
- "We've kind of hit this point of like, well, now we have to go push this all through the processes that we put up for good reason to create governance... moving the velocity of creation into the missing pieces of governance."
- "I can ask Claude... to create me a new website... and it might be beautiful... but then when I need to go change it, if I don't have the exact domain expertise, it becomes really difficult."
- "Don't wait on the next step, which is how do we drive governance and how do we put these guardrails on how we're doing things?"
Takeaways
- Bridge the Velocity-Governance Gap: The core bottleneck in enterprise AI adoption isn't generation speed, but approval speed. Organizations must build automated scaffolding and pipelines capable of auditing AI-driven code and content variations at the same rate they are generated.
- Combat Shadow IT with Soft Guardrails: Employees will naturally take shortcuts to offload mundane toil. Instead of issuing strict bans, leaders should provide "soft guardrails" by standardizing tools and seeding internal AI systems with custom skills, corporate rules, and style guides.
- Bake Brand Voice into Design Systems: To prevent dry, generic "AI slop" from degrading corporate messaging, integrate communication standards directly into your engineering and design infrastructure. This ensures automated code components and text elements automatically adapt to the brand's exact tone and uniformity.
- Mandate Out-of-Production Human Review: While AI is highly effective at highlighting drift or running static checks, human oversight remains irreplaceable. Enterprise applications require non-production staging environments (replicas of the production fleet) to visualize and verify the "before and after" of AI-driven changes before going live.
Find Amanda Razani on LinkedIn. https://www.linkedin.com/in/amanda-razani-990a7233/
Follow the FTA LinkedIn Page: https://www.linkedin.com/company/full-tech-ahead/
Visit the FTA website: https://fulltechahead.com/
Check out the Substack Channel: https://fulltechahead.substack.com/
Listen
Watch
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
SPEAKER_00 (00:19):
Hello and welcome to Full Tech Ahead.
I'm your host, Amanda Razzani,and I'm excited to be here today
with Chris Yates.
He is the SVP Product Design andEngineering at Pantheon.
How are you doing today?
SPEAKER_01 (00:32):
I'm doing great, Amanda.
Thanks for having me.
SPEAKER_00 (00:35):
Happy to have you on the show.
Can you share a little bit aboutthe services that Pantheon
provides?
SPEAKER_01 (00:41):
Yeah, so Pantheon is uh, as we say, where the web
works.
So uh we are focused on enablingorganizations to build uh and
ship on the web at scale, uh,which means both, you know, I
described it in threedimensions.
It means that we have customersthat need to build and run
(01:02):
business critical websites thatreceive sometimes you know tens
of millions of page views perhour in really spiky ways.
So that's the hard the verticalscale.
Uh we have customers that scalehorizontally, they don't have
one website, they have oftenhundreds or thousands of
websites and they have to figureout how to manage through that
chaos.
And the third dimension ispeople uh and the people that
(01:25):
form the web team, whetherthey're technologists, they're
editors, they're authors,they're designers, all have to
come and work together.
Uh and that is a uh aninteresting uh kind of ball, uh,
especially when you add in kindof our new coworkers that are
are uh toiling alongside us inthe form of uh LLMs and agents.
(01:46):
Uh, but we have uh you knowcustomers that have hundreds or
thousands of people involved intheir portfolios of websites.
So Pantheon's service is reallyoriented around serving the
whole web team uh regardless ofthat dimension of scale or
across all three.
SPEAKER_00 (02:03):
Awesome.
Well, we're gonna talk todayabout the implementation of AI
tools everywhere and the trustfactor that comes from that.
So my first question is justfrom your experience, where is
there a bottleneck when it comesto trusting AI and all these
(02:23):
tools?
And how is this affecting teams?
SPEAKER_01 (02:29):
I think the bottlenecks are just evolving uh
or emerging, especially forcustomers that are, you know,
especially like our customers,our users, the people that
create websites that run onPantheon are people that build
websites that matter to theirorganization, to their business.
And those customers often haverigorous approaches to security
(02:50):
and compliance and quality.
What we've seen with the kind ofthe this leap that AI tools have
made just in the last six monthsis like this incredible uh
improvement of the speed atwhich we can create things.
We can write code faster, we cancreate imagery and videos uh and
text faster.
(03:10):
But now we've kind of hit thispoint of like, well, now we have
to go push this all through theprocesses that we put up for
good reason to create governancearound the things that we do, to
ensure that we are uh keepingour our data and our our
customers' data secured toensure that we you know meet the
quality standards that weexpect, whether that's for you
(03:31):
know the the pros that uh weship as marketing copy or
editorial copy or you knowstories uh on a news website uh
or the performance of the sitethat uh we we code uh or design
together.
So that you know is is aninteresting challenge, kind of
moving the the velocity ofcreation into kind of the the
(03:54):
missing pieces of governance insome worlds, where we have to
figure out a way to reallycreate the scaffolding that
enables that that speed to kindof carry through to the actual
production realization of uh allof the magic that we get from
these new tools.
SPEAKER_00 (04:11):
Yeah, absolutely.
AI is speeding up, but how docompanies see what's actually
happening behind the scenes withAI?
How do they track that?
SPEAKER_01 (04:22):
It's it's been tough in the world of the web because
the you know, the way thatwebsites have been built for you
know 30 years uh you know reallyis multifaceted.
You have uh one element that isvery much technical and it is
driven by the code that youcreate.
The languages have shifted andand morphed over the years, but
(04:43):
you know, it's it's still codekind of creates the scaffolding
for how the site is going tolook, how the site is going to
behave, how the site is going toperform.
But then you have uh you knowcontent uh that hydrates that
code effectively.
And that comes through, youknow, in the old days it was uh
you know, you have a contentmanagement system that then
spits out content and code andcombines it and serves it on an
(05:05):
Apache web server.
Uh, I'm kind of dating myself atthat point, at this point.
But now it's you know, then wecan't went into uh full stack
kind of monolithic CMSs, Drupal,WordPress on the open source
side, site core, Adobe AM, andand and so on on the commercial
side of things.
And now we've kind of come fullcircle back to these decoupled
architectures with you know APIsdriving things like Next.js or
(05:29):
Astro.
The challenge across all ofthose things has been that there
are processes for looking intothe code.
Most organizations, mid-marketenterprise, are probably using
something like GitHub or GitLabor Bitbucket.
They've got checks that runwhenever you create a pull
request, whenever you make acommit.
A bunch of things are going tohappen.
And most of the time you cancatch, you know, uh, because
(05:50):
those tools are getting betterat doing static and dynamic
analysis, where there are bugs,where there are, you know,
things that, you know, secretsor keys or et cetera, things
that shouldn't be in the codethat might impact security or
performance.
But really, there hasn't been agreat scaffolding around how we
do the same with content and theintersection of content and
(06:10):
code.
Uh, so that's been a bigchallenge in in the industry as
we've just increased thevelocity of creation of all of
those things.
So that's one of the things thatI'm spending a lot of time on
personally, but also across thePantheon organization.
We're spending a lot of timebuilding that kind of next
generation of tools to go give adeep view into both the content
and the code base so that we canunderstand all of the changes,
(06:34):
whether they're human changes orthey're AI-driven changes, uh,
and get kind of the bird's eyeview within a site or across
many sites.
SPEAKER_00 (06:43):
From your experience, where are you seeing
the most negative impact orthings breaking down from people
relying too heavily on AIwithout that oversight?
SPEAKER_01 (06:54):
I think there's a there's always been this
tension, I think AI or no, uh,you know, historically between
kind of the the the needs of ofthe technology side and the
needs of the, I'll just call itthe business side in general.
You could call it it might bemarketing, it might be
editorial, whatever that is inin the organization, around kind
of like the control and thespeed and flexibility.
(07:16):
And I think what is happening isthere are more shortcuts, there
are there's more shadow IT,there are more things that
attach to the side oforganizations, uh, and you know,
enable some additional ease orspeed or kind of offloading of
toil from creation of of thingswith AI, but kind of come
through the side door versuscoming in the normal process by
(07:39):
which content or code changesget into the system.
And one of the other bigchallenges of that is inherent
with that is that understandingwhat the tools are doing becomes
increasingly challenging.
So I can ask Claude or ClaudeCode or uh Codex to create me a
new website, to create me a newpage, and it might be beautiful,
(08:01):
it might be mostly functional.
Um, but then when I need to gochange it, if I don't have the
exact domain expertise, itbecomes really difficult to go
say, well, I I actually wantedto do this or I want to make
sure it's not doing that.
That becomes super difficult.
So we really need to createscaffolding uh around how that
works, uh, so that when an AItool uh generates something for
(08:25):
us, it's not just vibing andtaking its, you know, all its
compendium of every amount ofhuman knowledge and using it to
shape a picture that we want,whether that picture is
literally an image or it's youknow a prototype of a web app or
a website.
We need something that allows usto kind of be on this kind of
common plane between you knowhow the a human user can
(08:46):
understand and then go andinteract and edit that content,
whether it's you know actualtext content, it's imagery, it's
it's a combination of thingsthat come into a web page, or
it's the code that sits belowit.
So I think that that scaffoldingand the kind of the challenge of
of things coming in the side uhin this as people adopt tools uh
very rapidly in organizations,um, you know, because it this
(09:10):
it's a very exciting time to bein technology, sometimes
terrifying time, but um a lot ofpeople are trying a lot of
different things at once.
SPEAKER_00 (09:19):
Yes.
Well, what advice would you havefor companies if they want to
install some guardrails or somesystems in place, make sure they
stay in control of the AIimplementation?
SPEAKER_01 (09:32):
Uh I mean, first of all, if if they're not already
implementing kind of a codereview pipeline that has
built-in kind of safeguards andchecks where uh you know
anything that's introduced intothe technical side, the code
side of your application,whether it's a monolith or it's
a decoupled or you know uhapplication, that you know,
(09:55):
you're applying those basicchecks, whether they're
available directly through atool like GitHub or you're
you're plugging them into othercode quality and vulnerability
detection tools, etc.
Uh the other side is a bit morechallenging today, but I think
there are there are safeguardsthat you can start to put in
place around uh how you think ofum kind of your organizational
(10:15):
governance.
So it's less of like the hardstops, but more of the soft
stops in terms of how you thinkabout standardization of tools.
So instead of picking you knowfour different prototyping
tools, let's you know say we'rewe're we're going to use this
one, you can build uh somescaffolding around how those uh
the code or the content isactually created.
(10:36):
Um, we've been leveraging customskills that we create, uh,
things like uh we've createdkind of Pantheon internal
skills, we've created anengineering codex that really
encompasses a lot of the uh thecustoms, a lot of the rules that
we use internally that we cankind of seed our AI tooling,
whether it's related toengineering work or it's related
(10:58):
to marketing work or it'srelated to content creation and
with kind of a rule set to startfrom.
Those are like soft guardrails.
Uh, and we're working on, interms of our product
development, on harderguardrails around how that the
actual integration of code andcontent work together to build
websites where you know we havea structure that enforces, you
(11:19):
know, uh basically a set ofcomponents that you can use that
can be assembled by humans orassembled by agents or uh
generative LLMs in order to kindof stay on that on the same
page, use this the the similar,the the this the common
substrate uh for humans andmachines to interact.
SPEAKER_00 (11:39):
I think another thing people are concerned about
is sort of there's a lot of theAI slop out there is the term I
hear a lot of yes and and verygeneric dry AI content.
So keeping brand voice and thevision of the brand, what advice
do you have when it comes tothat?
SPEAKER_01 (11:59):
That's one of the things that we've built into uh
it even just for our internaluse, we've actually built into
our design system.
So we have a really robustdesign system that our team
built to govern like howPantheon looks ass and works,
and it enables us to do thingsamazingly quickly.
We have designers that can spinup new prototypes uh in minutes
(12:21):
using uh generative AI tooling.
But we we also built into thatis some of those guardrails
around the corporatecommunication standards, around
the product language that weuse.
So we've baked that into thattool set.
So in addition to saying I, youknow, I can prompt uh Claude uh
or another uh tool to create a anew prototype that has really
(12:45):
tremendous fidelity andusability and matches the
standards of other products thatwe use that we are creating and
are in production at Pantheon.
We also ensure that like thelanguage that we put into that,
whether it's what a button saysor how we describe an alert, you
know, how how that actuallyreads, uh, you know, how it uh
is understood.
(13:05):
So we can create greateruniformity.
And that really is a tremendousshortcut to just the
productivity of being able to dothese things um and get those,
get these prototypes in thehands of our users uh to get
rapid feedback and and tightenthe cycle of development.
But it also applies to things onthe you know, more of the
go-to-market side of ourbusiness.
(13:26):
And I think that's also, youknow, regardless of what type of
business you are, I think that'san opportunity to codify some of
those standards about you knowhow you talk about your
organization, uh, what you say,what you don't say, create those
guidelines.
The next step really issomething that we're working on
as part of our product vision toenable that to just plug into
(13:49):
the content management uhtooling, to be able to have that
as basically a you know an agentthat might continuously work in
the background or a network ofagents driven by your own rules,
so driven by what theorganization defines as kind of
their standard for how contentis created or how images are
(14:09):
used or how media is presentedor how code is worked on, uh,
and enable that kind ofcontinuous improvement process
to go on in the background.
But you know, within the youknow, the other thing that I
think is really important uh asorganizations are considering
how to deploy AI createdessentially products or output
(14:30):
into production is do you have aprocess by which you can look at
this stuff first?
How do you put a human in theloop?
How do you ensure a qualityprocess?
All of these these things,whether they're automated checks
on your code or they'reautomated suggestions and
reviews uh of your content, uh,are wonderful, but I there's
really no substitute for saying,wow, that looks right.
(14:51):
I can see the before and aftervery clearly, which means you
have to have a structure forseeing the before you know what
what the the state, the futurestate will be.
So being able to have you know areplica of of production or
maybe several replicas ofproduction where you have
different work streams going on.
So having a system that that youcan kind of pre-flight all of
that, uh, all of the changesthat you're making, I think is
(15:14):
really critical as you're you'reintroducing a greater pace of
change into your organizationwith AI tools.
SPEAKER_00 (15:20):
Yeah, so having that human in the process is still
pretty critical.
SPEAKER_01 (15:24):
I I think so.
And like there's a lot of we canactually use AI to improve that
and call attention to the thingsthat have changed.
We can recognize the drift.
These tools are wonderful forthat, but you have to have a
place for that to run too.
So you have to have, you know,something outside of production
that you can look at to say,yep, that's right.
Uh, you know, what the AI hashas highlighted has changed, or
(15:45):
what the deterministic scripthas highlighted has changed, you
know, shows me what I need topay attention to.
And now I can say yes, you can'tjust do that in in production.
So you have to have this, uh,have a tool set uh that enables
you to have multiple versions ofof your of your site or even of
your fleet.
SPEAKER_00 (16:02):
Absolutely.
Well, if there was one keytakeaway you could leave our
audience today with, what wouldthat be?
SPEAKER_01 (16:08):
I think that we've gotten to the excitement stage
uh of this where like there'sreally rapid adoption.
I think we've we've turned thecorner in most organizations
from I don't know about this tooh my gosh, I need to use this
everywhere.
I think my my recommendation isdon't wait on the next step,
which is how do we you knowdrive governance and how do we
put these guardrails on uh howwe're we're doing things?
(16:31):
It doesn't mean necessarilyslowing them down, pulling, you
know, you know, putting on thebrakes on the whole thing, but
being thoughtful about uh youknow what is the human loop
process for evaluating uh thesechanges.
Uh, is there are there somefail-safes that we can
introduce?
Are there some ways that we cangenerate efficiency?
Because now I uh in mostorganizations that and other
(16:53):
with other technical technologyleaders I've talked to, there's
experimentation going oneverywhere in every single
department.
And I think you know, callinghaving those people come
together and and also have somekey leaders drive some
standardization is super helpfulbecause that will just drive
efficiency and safety andgovernance across the
organization.
SPEAKER_00 (17:12):
Wonderful.
Well, thank you so much forcoming on the show today and
sharing your insights.
SPEAKER_01 (17:17):
Absolutely.
This is great.
Thanks, Amanda.
SPEAKER_00 (17:19):
And thank you to our audience.
If you have any questions orcomments, of course, leave those
and I'll try to reply back.
And until the next podcast, havea wonderful week.
Hello and welcome to Full Tech Ahead.
I'm your host, Amanda Razzani,and I'm excited to be here today
with Chris Yates.
He is the SVP Product Design andEngineering at Pantheon.
How are you doing today?
SPEAKER_01 (00:32):
I'm doing great, Amanda.
Thanks for having me.
SPEAKER_00 (00:35):
Happy to have you on the show.
Can you share a little bit aboutthe services that Pantheon
provides?
SPEAKER_01 (00:41):
Yeah, so Pantheon is uh, as we say, where the web
works.
So uh we are focused on enablingorganizations to build uh and
ship on the web at scale, uh,which means both, you know, I
described it in threedimensions.
It means that we have customersthat need to build and run
(01:02):
business critical websites thatreceive sometimes you know tens
of millions of page views perhour in really spiky ways.
So that's the hard the verticalscale.
Uh we have customers that scalehorizontally, they don't have
one website, they have oftenhundreds or thousands of
websites and they have to figureout how to manage through that
chaos.
And the third dimension ispeople uh and the people that
(01:25):
form the web team, whetherthey're technologists, they're
editors, they're authors,they're designers, all have to
come and work together.
Uh and that is a uh aninteresting uh kind of ball, uh,
especially when you add in kindof our new coworkers that are
are uh toiling alongside us inthe form of uh LLMs and agents.
(01:46):
Uh, but we have uh you knowcustomers that have hundreds or
thousands of people involved intheir portfolios of websites.
So Pantheon's service is reallyoriented around serving the
whole web team uh regardless ofthat dimension of scale or
across all three.
SPEAKER_00 (02:03):
Awesome.
Well, we're gonna talk todayabout the implementation of AI
tools everywhere and the trustfactor that comes from that.
So my first question is justfrom your experience, where is
there a bottleneck when it comesto trusting AI and all these
(02:23):
tools?
And how is this affecting teams?
SPEAKER_01 (02:29):
I think the bottlenecks are just evolving uh
or emerging, especially forcustomers that are, you know,
especially like our customers,our users, the people that
create websites that run onPantheon are people that build
websites that matter to theirorganization, to their business.
And those customers often haverigorous approaches to security
(02:50):
and compliance and quality.
What we've seen with the kind ofthe this leap that AI tools have
made just in the last six monthsis like this incredible uh
improvement of the speed atwhich we can create things.
We can write code faster, we cancreate imagery and videos uh and
text faster.
(03:10):
But now we've kind of hit thispoint of like, well, now we have
to go push this all through theprocesses that we put up for
good reason to create governancearound the things that we do, to
ensure that we are uh keepingour our data and our our
customers' data secured toensure that we you know meet the
quality standards that weexpect, whether that's for you
(03:31):
know the the pros that uh weship as marketing copy or
editorial copy or you knowstories uh on a news website uh
or the performance of the sitethat uh we we code uh or design
together.
So that you know is is aninteresting challenge, kind of
moving the the velocity ofcreation into kind of the the
(03:54):
missing pieces of governance insome worlds, where we have to
figure out a way to reallycreate the scaffolding that
enables that that speed to kindof carry through to the actual
production realization of uh allof the magic that we get from
these new tools.
SPEAKER_00 (04:11):
Yeah, absolutely.
AI is speeding up, but how docompanies see what's actually
happening behind the scenes withAI?
How do they track that?
SPEAKER_01 (04:22):
It's it's been tough in the world of the web because
the you know, the way thatwebsites have been built for you
know 30 years uh you know reallyis multifaceted.
You have uh one element that isvery much technical and it is
driven by the code that youcreate.
The languages have shifted andand morphed over the years, but
(04:43):
you know, it's it's still codekind of creates the scaffolding
for how the site is going tolook, how the site is going to
behave, how the site is going toperform.
But then you have uh you knowcontent uh that hydrates that
code effectively.
And that comes through, youknow, in the old days it was uh
you know, you have a contentmanagement system that then
spits out content and code andcombines it and serves it on an
(05:05):
Apache web server.
Uh, I'm kind of dating myself atthat point, at this point.
But now it's you know, then wecan't went into uh full stack
kind of monolithic CMSs, Drupal,WordPress on the open source
side, site core, Adobe AM, andand and so on on the commercial
side of things.
And now we've kind of come fullcircle back to these decoupled
architectures with you know APIsdriving things like Next.js or
(05:29):
Astro.
The challenge across all ofthose things has been that there
are processes for looking intothe code.
Most organizations, mid-marketenterprise, are probably using
something like GitHub or GitLabor Bitbucket.
They've got checks that runwhenever you create a pull
request, whenever you make acommit.
A bunch of things are going tohappen.
And most of the time you cancatch, you know, uh, because
(05:50):
those tools are getting betterat doing static and dynamic
analysis, where there are bugs,where there are, you know,
things that, you know, secretsor keys or et cetera, things
that shouldn't be in the codethat might impact security or
performance.
But really, there hasn't been agreat scaffolding around how we
do the same with content and theintersection of content and
(06:10):
code.
Uh, so that's been a bigchallenge in in the industry as
we've just increased thevelocity of creation of all of
those things.
So that's one of the things thatI'm spending a lot of time on
personally, but also across thePantheon organization.
We're spending a lot of timebuilding that kind of next
generation of tools to go give adeep view into both the content
and the code base so that we canunderstand all of the changes,
(06:34):
whether they're human changes orthey're AI-driven changes, uh,
and get kind of the bird's eyeview within a site or across
many sites.
SPEAKER_00 (06:43):
From your experience, where are you seeing
the most negative impact orthings breaking down from people
relying too heavily on AIwithout that oversight?
SPEAKER_01 (06:54):
I think there's a there's always been this
tension, I think AI or no, uh,you know, historically between
kind of the the the needs of ofthe technology side and the
needs of the, I'll just call itthe business side in general.
You could call it it might bemarketing, it might be
editorial, whatever that is inin the organization, around kind
of like the control and thespeed and flexibility.
(07:16):
And I think what is happening isthere are more shortcuts, there
are there's more shadow IT,there are more things that
attach to the side oforganizations, uh, and you know,
enable some additional ease orspeed or kind of offloading of
toil from creation of of thingswith AI, but kind of come
through the side door versuscoming in the normal process by
(07:39):
which content or code changesget into the system.
And one of the other bigchallenges of that is inherent
with that is that understandingwhat the tools are doing becomes
increasingly challenging.
So I can ask Claude or ClaudeCode or uh Codex to create me a
new website, to create me a newpage, and it might be beautiful,
(08:01):
it might be mostly functional.
Um, but then when I need to gochange it, if I don't have the
exact domain expertise, itbecomes really difficult to go
say, well, I I actually wantedto do this or I want to make
sure it's not doing that.
That becomes super difficult.
So we really need to createscaffolding uh around how that
works, uh, so that when an AItool uh generates something for
(08:25):
us, it's not just vibing andtaking its, you know, all its
compendium of every amount ofhuman knowledge and using it to
shape a picture that we want,whether that picture is
literally an image or it's youknow a prototype of a web app or
a website.
We need something that allows usto kind of be on this kind of
common plane between you knowhow the a human user can
(08:46):
understand and then go andinteract and edit that content,
whether it's you know actualtext content, it's imagery, it's
it's a combination of thingsthat come into a web page, or
it's the code that sits belowit.
So I think that that scaffoldingand the kind of the challenge of
of things coming in the side uhin this as people adopt tools uh
very rapidly in organizations,um, you know, because it this
(09:10):
it's a very exciting time to bein technology, sometimes
terrifying time, but um a lot ofpeople are trying a lot of
different things at once.
SPEAKER_00 (09:19):
Yes.
Well, what advice would you havefor companies if they want to
install some guardrails or somesystems in place, make sure they
stay in control of the AIimplementation?
SPEAKER_01 (09:32):
Uh I mean, first of all, if if they're not already
implementing kind of a codereview pipeline that has
built-in kind of safeguards andchecks where uh you know
anything that's introduced intothe technical side, the code
side of your application,whether it's a monolith or it's
a decoupled or you know uhapplication, that you know,
(09:55):
you're applying those basicchecks, whether they're
available directly through atool like GitHub or you're
you're plugging them into othercode quality and vulnerability
detection tools, etc.
Uh the other side is a bit morechallenging today, but I think
there are there are safeguardsthat you can start to put in
place around uh how you think ofum kind of your organizational
(10:15):
governance.
So it's less of like the hardstops, but more of the soft
stops in terms of how you thinkabout standardization of tools.
So instead of picking you knowfour different prototyping
tools, let's you know say we'rewe're we're going to use this
one, you can build uh somescaffolding around how those uh
the code or the content isactually created.
(10:36):
Um, we've been leveraging customskills that we create, uh,
things like uh we've createdkind of Pantheon internal
skills, we've created anengineering codex that really
encompasses a lot of the uh thecustoms, a lot of the rules that
we use internally that we cankind of seed our AI tooling,
whether it's related toengineering work or it's related
(10:58):
to marketing work or it'srelated to content creation and
with kind of a rule set to startfrom.
Those are like soft guardrails.
Uh, and we're working on, interms of our product
development, on harderguardrails around how that the
actual integration of code andcontent work together to build
websites where you know we havea structure that enforces, you
(11:19):
know, uh basically a set ofcomponents that you can use that
can be assembled by humans orassembled by agents or uh
generative LLMs in order to kindof stay on that on the same
page, use this the the similar,the the this the common
substrate uh for humans andmachines to interact.
SPEAKER_00 (11:39):
I think another thing people are concerned about
is sort of there's a lot of theAI slop out there is the term I
hear a lot of yes and and verygeneric dry AI content.
So keeping brand voice and thevision of the brand, what advice
do you have when it comes tothat?
SPEAKER_01 (11:59):
That's one of the things that we've built into uh
it even just for our internaluse, we've actually built into
our design system.
So we have a really robustdesign system that our team
built to govern like howPantheon looks ass and works,
and it enables us to do thingsamazingly quickly.
We have designers that can spinup new prototypes uh in minutes
(12:21):
using uh generative AI tooling.
But we we also built into thatis some of those guardrails
around the corporatecommunication standards, around
the product language that weuse.
So we've baked that into thattool set.
So in addition to saying I, youknow, I can prompt uh Claude uh
or another uh tool to create a anew prototype that has really
(12:45):
tremendous fidelity andusability and matches the
standards of other products thatwe use that we are creating and
are in production at Pantheon.
We also ensure that like thelanguage that we put into that,
whether it's what a button saysor how we describe an alert, you
know, how how that actuallyreads, uh, you know, how it uh
is understood.
(13:05):
So we can create greateruniformity.
And that really is a tremendousshortcut to just the
productivity of being able to dothese things um and get those,
get these prototypes in thehands of our users uh to get
rapid feedback and and tightenthe cycle of development.
But it also applies to things onthe you know, more of the
go-to-market side of ourbusiness.
(13:26):
And I think that's also, youknow, regardless of what type of
business you are, I think that'san opportunity to codify some of
those standards about you knowhow you talk about your
organization, uh, what you say,what you don't say, create those
guidelines.
The next step really issomething that we're working on
as part of our product vision toenable that to just plug into
(13:49):
the content management uhtooling, to be able to have that
as basically a you know an agentthat might continuously work in
the background or a network ofagents driven by your own rules,
so driven by what theorganization defines as kind of
their standard for how contentis created or how images are
(14:09):
used or how media is presentedor how code is worked on, uh,
and enable that kind ofcontinuous improvement process
to go on in the background.
But you know, within the youknow, the other thing that I
think is really important uh asorganizations are considering
how to deploy AI createdessentially products or output
(14:30):
into production is do you have aprocess by which you can look at
this stuff first?
How do you put a human in theloop?
How do you ensure a qualityprocess?
All of these these things,whether they're automated checks
on your code or they'reautomated suggestions and
reviews uh of your content, uh,are wonderful, but I there's
really no substitute for saying,wow, that looks right.
(14:51):
I can see the before and aftervery clearly, which means you
have to have a structure forseeing the before you know what
what the the state, the futurestate will be.
So being able to have you know areplica of of production or
maybe several replicas ofproduction where you have
different work streams going on.
So having a system that that youcan kind of pre-flight all of
that, uh, all of the changesthat you're making, I think is
(15:14):
really critical as you're you'reintroducing a greater pace of
change into your organizationwith AI tools.
SPEAKER_00 (15:20):
Yeah, so having that human in the process is still
pretty critical.
SPEAKER_01 (15:24):
I I think so.
And like there's a lot of we canactually use AI to improve that
and call attention to the thingsthat have changed.
We can recognize the drift.
These tools are wonderful forthat, but you have to have a
place for that to run too.
So you have to have, you know,something outside of production
that you can look at to say,yep, that's right.
Uh, you know, what the AI hashas highlighted has changed, or
(15:45):
what the deterministic scripthas highlighted has changed, you
know, shows me what I need topay attention to.
And now I can say yes, you can'tjust do that in in production.
So you have to have this, uh,have a tool set uh that enables
you to have multiple versions ofof your of your site or even of
your fleet.
SPEAKER_00 (16:02):
Absolutely.
Well, if there was one keytakeaway you could leave our
audience today with, what wouldthat be?
SPEAKER_01 (16:08):
I think that we've gotten to the excitement stage
uh of this where like there'sreally rapid adoption.
I think we've we've turned thecorner in most organizations
from I don't know about this tooh my gosh, I need to use this
everywhere.
I think my my recommendation isdon't wait on the next step,
which is how do we you knowdrive governance and how do we
put these guardrails on uh howwe're we're doing things?
(16:31):
It doesn't mean necessarilyslowing them down, pulling, you
know, you know, putting on thebrakes on the whole thing, but
being thoughtful about uh youknow what is the human loop
process for evaluating uh thesechanges.
Uh, is there are there somefail-safes that we can
introduce?
Are there some ways that we cangenerate efficiency?
Because now I uh in mostorganizations that and other
(16:53):
with other technical technologyleaders I've talked to, there's
experimentation going oneverywhere in every single
department.
And I think you know, callinghaving those people come
together and and also have somekey leaders drive some
standardization is super helpfulbecause that will just drive
efficiency and safety andgovernance across the
organization.
SPEAKER_00 (17:12):
Wonderful.
Well, thank you so much forcoming on the show today and
sharing your insights.
SPEAKER_01 (17:17):
Absolutely.
This is great.
Thanks, Amanda.
SPEAKER_00 (17:19):
And thank you to our audience.
If you have any questions orcomments, of course, leave those
and I'll try to reply back.
And until the next podcast, havea wonderful week.
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Crime Junkie
Does hearing about a true crime case always leave you scouring the internet for the truth behind the story? Dive into your next mystery with Crime Junkie. Every Monday, join your host Ashley Flowers as she unravels all the details of infamous and underreported true crime cases with her best friend Brit Prawat. From cold cases to missing persons and heroes in our community who seek justice, Crime Junkie is your destination for theories and stories you won’t hear anywhere else. Whether you're a seasoned true crime enthusiast or new to the genre, you'll find yourself on the edge of your seat awaiting a new episode every Monday. If you can never get enough true crime... Congratulations, you’ve found your people. Follow to join a community of Crime Junkies! Crime Junkie is presented by Audiochuck Media Company.
NFL Daily with Gregg Rosenthal
Gregg Rosenthal and a rotating crew of elite NFL Media co-hosts, including Patrick Claybon, Colleen Wolfe, Steve Wyche, Nick Shook and Jourdan Rodrigue of The Athletic get you caught up daily on all the NFL news and analysis you need to be smarter and funnier than your friends.