April 7, 2025 • 33 mins
Key Topics Covered
- Why AI is still in its “stupidest” version—and why that’s exciting
- The difference between AI as a collaborator vs. a replacement
- How knowledge graphs will define the next generation of software
- The danger of adopting AI without cleaning up the process first
- Why Comcast’s Xfinity platform is one to watch
- Real-world advice for tech leaders preparing for AI integration
- From leadership scar tissue to startup pivots: the human side of innovation
Mentioned in This Episode
- AI Circle – A community for emerging AI leaders and thinkers
- Comcast Xfinity – A model for next-gen customer intelligence
- Aduna & Camara – Telecom’s move toward open, intelligent APIs
- CENX & Ericsson – How a technical pivot led to acquisition
- Knowledge graphs – The new competitive advantage in AI
Thomas’ Wisdom Highlights
“Every day, you’re working with the dumbest version of AI you’ll ever use.”
“Before you apply AI, ask: why are we doing this in the first place?”
“AI will supercharge the smart. But if your processes are broken, AI only accelerates the chaos.”
“Sometimes, the biggest hurdle to growth is getting out of your own ego’s way.”
About the Guest
Thomas Elliott is a disruptive technology leader and AI strategist. He’s led engineering teams through major scaling, cloud-native transformations, and successful exits. Currently Principal at Datanised and a founding member of AI Circle, Thomas advises global organizations on how to adopt AI responsibly while fostering innovation and continuous improvement. He’s also passionate about helping leadership teams rethink their operating models to meet the demands of the AI era.
Connect with Thomas Elliott
LinkedIn: www.linkedin.com/in/Elliott-cloudnative
Stay Connected
Enjoyed this episode? Make sure to subscribe and stay updated for more insightful conversations with inspiring guests.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:02):
Welcome
to IT Intelligence,
where we explore the latest trends and ways
to leverage artificial intelligence
in the IT industry.
Learn from innovators and IT leaders
and reveal the opportunities and risks of the
AI era.
And now,
(00:23):
here is your host,
Barnes Lam.
Hello. Today's
guest is Thomas Elliott,
a disruptive tech leader and trailblazer in artificial
intelligence,
cloud native
systems and scalable innovation.
With over two decades of experience in leading
(00:44):
startups
through exclusive growth and major acquisitions,
Thomas has become the go to expert in
building resilient,
high performing
engineering teams
while embracing the bleeding edge of technology.
From self heating networks to real time fault
detection,
(01:04):
he has been shaping the future
of tech before it had a name. He
is currently advising global AI and data strategy
as principal
at DataNash,
while also fostering executive tech committees
through Filio.
If you are curious about the real world
impact of AI,
(01:25):
team growing strategies,
what are coming beyond 02/2025?
This episode is one you do not want
to miss.
Welcome, Thomas. Thank you for joining the show
today. How are you doing? Thank you. Quite
the introduction. I'm doing quite well. Thank you.
Thank you again.
Let's start. You have led teams through massive
(01:46):
scaling and multiple acquisitions.
How has artificial intelligence
influenced your leadership and team strategy? Oh. AI
has been along for the ride for a
long time for me. And
a lot of people
see AI and equate it to generative AI.
My AI journey
started with products that we were bringing the
(02:08):
market back in 02/2004
at a security company called Entrust,
where we introduced
fuzzy logic mail filtering,
and then later real time fraud detection. So
how has it influenced my
perspective on leadership? That's a really, really good
question because it's meant that my focus
has been on driving results through innovative tooling.
(02:31):
I've always wanted to make sure that the
tools that my team had access to were
as innovative
as the tools that we were pitching and
bringing to market. So being a leader of
service teams, of field services teams,
where
we're selling this innovation
and
we are pitching this vision of the future,
(02:52):
being able to
keep up with our product and being able
to introduce the collaboration
and
the capabilities
of what makes high functioning teams
work at that next level
has always been
something that has been meaningful to me in
terms of the technologies that I would gravitate
to. Now generative AI is really
(03:15):
opening the bounds of accelerating
the capabilities
of individuals.
And we're just the beginning stage of, you
know, of being able to understand what this
means
from a leadership perspective
of what new shapes our teams are going
to take.
Great. Thank you. This is really good for
(03:35):
Satway for our next question.
Obviously, you just mentioned, alluded to earlier, you
have been working with AI since early two
thousands.
What excites you the most about AI in
02/2025
and the decades ahead? You mentioned about generative
AI and all these different things that we
just started. Can you give to our audience
a taste of what you see from your
perspective? There's a great quote that I hear
(03:57):
repeated frequently, and that is, when you're working
with this technology right now, every day you
wake up, I am working with the stupidest
version of this technology
that I will ever work with. Interesting.
So
despite
the capability
and the acceleration
potential
that's baked in today,
(04:18):
This is really the least capable
technology
that I'm gonna work with in my lifetime.
Interesting. Wow. And it I guess this comes
from a telecom notebook where I have, like,
a ten year perspective
of what to expect and and
where
skate to where the puck is going. Mhmm.
Yes.
A fellow Canadian. Yes.
(04:39):
The great Canadian.
And that is
it really becomes
there's an element of cognitive incoherence
Mhmm. Of being able to understand
a future self
and what that reality is. The only thing
that we can do is look at patterns
that are evolving now
and try to predict
(05:00):
the patterns that will spin
on that trajectory.
So
we think of we think of
the improvements
in cognitive capabilities of AI. Right now, working
with transformers,
we're
seeing artificial intelligence
that is able to
(05:20):
begin
faking out reasoning,
where it's a large language model and it's
trying to
predict
a conversation
or what makes sense
in terms of a response. And we all
know that AI can get ahead of itself
or
in the absence of facts and its model,
(05:40):
completely hallucinate. I like that one.
But it's in the future
with how models
and
new versions of intelligence will evolve,
we'll be able to
get beyond just the frontal cortex
of the brain, which is just a reasoning
engine, and be able to
(06:01):
look at additional theories of mind where we're
looking at short term,
long term memory, how that interplay,
how that ends up being self referential
in being able to evaluate
responses.
And then, of course,
implementation of doubt.
Doubt is a highly
functional,
(06:22):
self referential cognitive model
that AI doesn't have
yet.
And it's part of what makes
a model
of what would make a model
questioned its response before speaking.
Wow. That's something that we've evolved ourselves.
Right. So
this is the thing that blows your mind
(06:43):
is that even if we get to the
point where we have been able to orchestrate
that or or program this and create a
model for it, there's very little in terms
of observability
that will allow us to understand
why it's making that decision. It will be
not purposely obfuscated
from us. It will be part of a
(07:03):
thought model that is
core to the intelligence itself. It's kinda like,
why would I ask Barnes?
How did he come up with that? And
there's, like, a multi factored response. And ultimately,
it's just it's this response felt right. But
when you say felt right,
so there's a certain part
of your
decision making
(07:24):
rests on human
rather than rests on artificial intelligence. Currently,
absolutely, yes. The decision making
is assisted by artificial intelligence. Right? Okay. Right.
Right. It's about having another member of your
team. Right? Mhmm. And right now, that team
member is at a point in its, I
(07:45):
guess, cognitive capacity. Mhmm. That's somewhat limited.
Mhmm. Right?
It can
summarize
really, really well. It can even
do some fantastic
correlation.
Mhmm. But you give it a lot of
facts. You give it a lot of interconnected
stories. And it has
well, let's be honest.
(08:06):
It has very similar
challenges
that a wet neural network that our brain
has. Mhmm. In that, it can get a
little mixed up.
Right. Connections.
It's a concept of entropy,
where you have connections being made where they
should not be made, and then you have
connections that are being made where they maybe
were not needed.
(08:27):
Right.
I understand. Understand. That's just really, really good.
I never thought that way at all.
This is I think my audience would love
to hear from you about these questions, I
think.
What do you see the biggest buying spots
companies still have when it come to implementing
AI at scale?
(08:47):
Pick one.
You know, where it's Yeah.
There is
I think that
moving
into
there's a long tail with agency right now.
And everybody's hot for AI agents.
We're so very early
in the beginning of that. But it's where
(09:09):
you end up with multiple
Think of it as like microservices.
Right. Right? Mhmm. When we launched microservices,
not everybody really created a microservice very well.
We're now in the stage of AI where
people are building an agent
and
sometimes
it's this could have been solved with a
(09:30):
regular expression.
Right? Yes. Or it could have been solved
with a bash script. Mhmm. That compute would
be a hell of a lot cheaper. Right.
But when you really branch into
the elements of agency of being able to
do correlation of data Mhmm. And
(09:50):
be able to
have,
user profiling
that's connected with a correlation engine.
I think that is
going to be one of the most valuable
elements
of what telecom companies can use of what
basically any customer
that has, you know, hundreds of thousands of
(10:11):
of customers can do. And they want to
be able to understand better, how is it
that customers are using my software?
What are their challenges with using my software?
And what is my software missing that my
customers are needing?
And AI can be an element in helping
us
understand what that is, prioritizing
(10:33):
what needs to be delivered,
and helping product managers
build the future of what it is that
they are building as a product,
as a collaborator,
not as a replacement for what it is
they already have.
As an example,
one company that I
see great things for is
what Comcast is doing with their
(10:55):
Xfinity product.
What you what is singularly unique about what
Comcast that Comcast has done
is that they are pulling together this product
that's not just a network in a box,
but
a full
customer profile
graph management system.
(11:16):
Mhmm. And
that profile
of their customers'
product use engagement,
family demarcations,
and
who has access to what and when.
And all of that
forms
an incredibly valuable knowledge graph Mhmm. That will
(11:36):
eventually be
incredibly useful for artificial intelligence.
Anybody who's building knowledge graph of their
of their customer activity
and
of how it is their product is is
being used Mhmm. Will have a leg up
on
the next generation of software that's being built.
(11:57):
Awesome. That's really the important point. Thank you.
You have worked with
everything from real time fault detection to self
healing networks.
Which use case do you think will define
AI's
in mainstream
moment in enterprise tech? You know, from coming
from telecom,
(12:17):
I think it's a really great
space to have been exposed to AI
Because telecom enterprises have been running AI innovation
since the seventies and eighties.
Right? If we go back to DMS
and early electronic switching,
(12:38):
there was
the early stages
of creating
rich data sources
for
automating heuristics
and
doing
predictive fault or anticipating
network faults.
And everything that's been happening since then has
been rolling forward
(12:59):
ease
easing in tools of
automation. We went through phases of monolithic software
of of
where you had bounded
hardware and software being shipped. And then we
went to
decoupled hardware and software with monolithic
software elements. And then we went into cloud
(13:20):
native software where we
decoupled the software and introduced microservices
that are interoperating.
All of these patterns
move us in the direction
of
greater senses of scale,
greater senses of programmability
of services that we're creating.
So that the elements that I see where
(13:41):
AI comes to play
is not that AI ends up being an
overriding
or super controlling element,
but
as a application collaborator.
And
that's where we have
efforts that are going on right now,
where we're introducing
(14:02):
common API frameworks.
In telecom, we have efforts like
Aduna and,
Kamara,
which is the creation of
open
intelligent APIs.
And these are APIs that you could have
applications interact with. You could have AI agents
(14:23):
interact with. So it's about
building out on those patterns.
You're
leaning into openness
where
you're allowing
communities of developers
to participate
with the interfaces that you're developing.
And then also
embracing open source technologies.
(14:44):
And all of this intersects with AI because
if you build good interfaces, if you build
good scalability, there's nothing that prevents
AI from becoming a consumer of those interfaces
and interoperating
with those interfaces.
So that those applications don't just empower other
applications,
but they empower
(15:05):
additional intelligences.
Mhmm. That's a key part. That's absolutely. Thank
you. You have helped companies
reduce delivery complexity by up to 60%.
How much of that was due to AI
versus leadership and process?
Oh, man.
You really hit the nail on the button
there because you don't get to one without
(15:26):
the other. Talking about processes brings me back
to the trauma and scar tissue from, like,
ISO
certification.
So I tend to avoid using processes
and I turn to lean into ways of
working.
It may mean the same thing, but I
just feel ways of working is just a
better way of describing what we're getting to.
(15:46):
And that is, if you establish good patterns
for being able to work together as a
team and to collaborate with other teams,
and those
ways of working are predictable,
documentable,
and can operate at scale,
then you have the capability of being able
to add in
(16:06):
artificial intelligence
as a collaborator and accelerator.
If you don't have that, I'll just lean
into it. It's garbage in, garbage out. All
the only thing that AI will do in
that case
is put you on a turnpike to failure.
Oh, direct. We will do. You may get
to a failure point very, very quickly. Quickly.
(16:27):
Faster. Right. Right.
You are a founding member of AI Circle,
and I really encourage our audience to check
it out. What trends are emerging in that
community
that others should know about? The AI circle
community is it's been a really great organization
for people that have this keen interest in
(16:47):
advancing
artificial intelligence
and want to be able to have a
community
where you have like minded people
that are able to, you know, throw spaghetti
on the wall. Mhmm. I like that.
Right? Yeah. Yeah. It's messy. It's insightful.
It's camaraderie.
And most importantly,
(17:08):
it creates a sense of community
of people that are speaking the same language.
Mhmm.
I can't really point to one single thing
that's coming out as
as a trend, but I'd say the common
focus
of people in AI circle is looking at
practical and predictable uses
(17:28):
of artificial intelligence
that will stand up in production,
that are able to demonstrate the best that
the technology
can bring to a solution, and also be
able
to
show how it
accelerates
our capabilities
as human beings.
Well, so Because it's
(17:48):
the one thing that's clear in this community
is that nobody is looking at these solutions
as
a replacement for human effort.
Mhmm.
It's all about
the acceleration
of our efforts.
Perfect. Thank you. With your global
work from Poland to Brazil to US,
(18:09):
how do you see AI adoption
differing across regions
or cultures
in a general term?
You're pulling the hard ones.
It's
I feel somewhat disconnected
over the past
six or so months,
having just started recently working with
global teams again. But I think one of
(18:31):
the things that's universal
is
that no matter where it is that you
go,
if you're connecting to people elsewhere in the
world,
they all share the same concerns over AI.
There is a very real concern of,
will this replace me? And
will this
efficiency result in
(18:52):
reductions?
My sense of optimism
is all around
the empowerment of individuals
to be able to do more, to be
able to understand more, and to be able
to move their efforts
forward
with a greater understanding of why
they're doing those things,
of how they're going to be doing those
(19:12):
things.
And
I think that
if we expand
AI's influence
beyond just
large language models and
responses out of GPTs.
We see the real
capabilities
being
going back to the the
some of the very early origins
(19:33):
of AI, which is natural language processing.
And you had
IVRs
explore this back in the eighties
with voice activated,
menuing systems
on IVRs.
Right. And now we have
universal translators on our iPhones.
Mhmm. Where I can go to Japan or
(19:55):
Thailand or Russia
and be able to point my phone at
a at a sign. And it renders
a foreign alphabet
into
decipherable English.
Mhmm. Or at least usable English. Yes. Usable.
That's right. I think any kind of technology
like that makes us feel less like aliens
in our own world.
(20:16):
Absolutely.
Thank you. I promise, I mean, I can
go on for another hour, but I don't
I don't think we have that. But, you
know, I just I already like to my
audience, I have questions. I already go through
half of my questions.
But anyway,
let's go on. What advice would you give
to a tech leaders
preparing for AI driven future,
(20:36):
especially though just now starting to embrace it?
My advice is
when you're creating applications
or when you are looking at your enterprise,
look first as to
what you're actually doing.
And
do not be afraid to answer the question
of why are we doing this?
(20:57):
I think one of the key patterns
of embracing
an AI first way of thinking
is
one of looking at a
process that begins
to simplify
complexity.
Because once you are able to understand what
you're doing
and
(21:18):
you can put
a workflow or documentation or repeatable success around
ways of working,
That's where you can introduce these accelerators.
If we follow the patterns of what's happened
in telecom,
you've
moved from these
complex
service orchestrations
and complex
(21:39):
delivery
of AI
and increasing use of AI and all of
these APIs throughout the network. We're now at
a point where
I'm not saying the network at this point
is self aware,
but it finally has enough self respect to
clean up after itself.
And that is really
one of the key elements of what AI
(22:00):
enables.
Cool. I promise. Not too much longer. I
got a couple more questions and and that's
it.
Depending on your career, what do you consider
your most significant achievement?
And what lessons did you learn from it
that you continues to apply today? You know,
I go back and think about my first
trip into leadership. I was
(22:21):
beyond
fortunate
in that I was a technical resource that
operated the peak of the organization,
top level contributor, the kind of person that
nobody wants to become a manager because they're
gonna lose their top contributor.
Interesting. Yes. I agree. I made that jump
and
I had a leadership team that did not
(22:43):
make it easy for me.
There was a lot of scar tissue that
was developed in that transition
and they held me
to a very, very high standard.
Everything that I learned
Well, not everything, but a lot of the
foundations of what it is that I learned
about leadership, about people, about
(23:04):
leading others,
and making sure that others can succeed.
I learned from
the people that took me under their wing
and
explained the importance
of being able to
not just
manage people,
but truly
become
(23:24):
a valued part of
the careers of others.
Being invested
in where people are taking their career,
how they're contributing as an individual.
A sudden realization is that a lot of
the core fundamentals of what it was that
I've learned there
go back to establishing some of the core
(23:44):
fundamentals of what is is I expect of
artificial intelligence.
How can I develop this intelligence
to be better than it is today? That's
a really good
point that you just made.
Thank you. Reflecting on your career,
what do you consider your most significant achievement
and
(24:05):
from your perspective
that you can apply today? At my Canadian
start up, Scenics,
there was a period
of time there
where
a new leader was introduced into the organization.
And
I had
what, at the time, I had perceived as
a personal setback.
And it basically felt like I
(24:28):
was taking a back seat,
a backward,
regression in my career. And over a period
of six months,
I really
leaned into that
and used it as a pivot in my
career
to become
highly specialized in distributed databases and cloud native
ways of working. Mhmm.
(24:48):
Ultimately, that level of technical expertise,
it steered the company in the direction
that it needed to be. Because
as much as I went in there and
did a reset of my career,
I also ended up delivering a needed reset
when a pivot
to the entire organization
(25:08):
because it ended up changing
the patterns and how it is that we
delivered software,
fully embracing cloud native automation.
It reduced
the level of effort for deploying our software
from, like,
weeks
to
a couple of hours if you count if
if you count downloading all of the container
images. Mhmm.
(25:29):
And then patching was a trivial exercise that
previously was, like, a full day's effort. We
reduced it to, like, five minutes. Wow.
So the infrastructure that we built up to
lead us to that direction to improve our
capabilities of delivery
ultimately ended up being the differentiator
and a reason
(25:49):
for acquisition
because we were able to demonstrate
a at scale
cloud native
product operating at a tier one carrier
that was
averse to change.
And
we were the one of the first cloud
native apps to operate
at this carrier,
and they were willing to go to bat
(26:11):
in a public press release
to talk about
how our product and our ways of delivering
the product
is changing
the way in which they're operating data centers.
So overall, I'd I'd say that that would
probably be one of the most significant
contributions that I've made in my career
of
being able to take something that looked like
(26:32):
a career regression
and turning it into
a success for the broader organization and then
carrying those patterns on into Ericsson who acquired
us. Mhmm.
Wow.
Amazing. I mean, obviously, you you come a
long way.
And, you know, your journey so far,
I'm sure that any successful person like yourself,
(26:54):
expertise like yourself, there are some challenging moment,
some hurdles that you have to overcome
to succeed in who you are today. What
are those hurdles?
Getting out of the way of my own
ego. I love it.
There is,
when you're going through a transformative
period of your career,
there is
(27:15):
an element, you know, I don't want to
get too hippie on you, but
there is an element of ego death
where you are surrendering
some of the ideas
of the foundations of what it is you're
believing in. And
whether or not functionally,
they can play a part in
(27:37):
where you need to be able to take
yourself.
Being in a transformative
period of your career means that you have
to be able to be open
to
letting go
of that
and to embrace new patterns and new new
ways
new ways of working, new ways of thinking.
Mhmm.
And it's not to do so just because
(27:57):
it's the new way of thinking.
It's because
you're either embracing patterns that are keeping you
in place or you're embracing patterns that are
moving you forward. I like that one.
Never put it that way. I like it.
I really like it.
Thank you.
I like it a lot, actually. What are
three tips that make you successful in what
(28:18):
you do?
I have when I'm interested in something, I'm
really, really interested at it. And I dive
deep into it. And
when you're in a leadership position and you
can understand something deeply,
it doesn't mean that you
are necessarily going to get your hands dirty
and start coding. But you're going to have
(28:38):
a better appreciation for what the people that
you are leading
are doing. Mhmm. And you can look at
their patterns of working with greater clarity.
And
you can see,
sometimes,
another
or better way or streamlined way of reducing
(28:59):
complexity,
of simplifying ways of working
with a full appreciation of what it means
to be working with the technology
at hand. Mhmm.
Wow.
I now I understand that
working with artificial intelligence really changing the way
you look at things so so deep differently.
That's amazing.
You accomplished so much in what
(29:21):
in your career. What are your goals and
aspiration in 02/2025
and beyond?
So, you know, with
my some odd years
behind me,
I'm now at the point where I am
looking to be able to get into advisory
and consulting roles, where I'm going to be
(29:42):
able to take the things that I've learned
and the patterns
that
have served me
and
my vision of
how things could be working. And I wanna
be able to begin sharing that process, that
way of working with others
so that I can
help them enable
their teams,
their styles of leadership
(30:03):
on
being able to
look at alternative trajectories,
look at
adding
continuous improvement loops into their workflow,
being able to
look at
how it is that their that their teams
can
be served better by a different style of
(30:24):
leadership or
levels of focus to their organization.
Cool. Last question.
If you could give yourself some tips for
your younger
self,
what would that be? Be less afraid of
letting go. Be less afraid of letting go.
Okay.
Be less afraid of letting go. There's a
difference between
(30:44):
being proud of your accomplishments
and
being able to
believe
in
certain patterns.
But there comes a time where
sometimes those patterns may not
be as helpful as they were.
Mhmm. And you need to be able to
let go and embrace these new patterns,
(31:06):
these new ways of working. And that applies
personally,
as much as it applies to
patterns in
organizations
that may have grown out of necessity,
but now
are collapsing under the weight
of legacy.
So that openness,
that lack of fear
(31:27):
to be able to embrace new patterns is
so it's so transformational,
and it's so
it's become so much
behind,
I guess,
my own personal operating system.
I like that. I like that one. Yes.
Thank you, Thomas. It has been wonderful
to having you on the show.
(31:47):
That's a wrap on today's conversation with Thomas
Elliott,
someone who truly
exemplifies what it means to lead with both
vision
and execution in the AI era. Whether it's
building teams that scale with soul, redefining what's
possible in cloud native infrastructure,
or advising the next wave of AI leaders,
(32:08):
Thomas is a force
at the intersection of technology
and leadership.
If you enjoyed this episode, please like, follow,
and share with someone thinking about how AI
will shape the next decade.
And don't forget to connect with Thomas on
LinkedIn
or
at a website, which I will be
listed.
(32:29):
Until next time, keep learning, keep building, keep
pushing boundaries.
See you next time on IT Intelligent Podcast.
Goodbye for now.
Thanks for joining us for IT Intelligence.
If you enjoyed this episode,
subscribe and leave us a review so we
can reach more tech heads and industry thought
(32:50):
leaders
like you.
For more information,
extended show notes, and additional resources on today's
topic, visit ITIntelligence.co.
We hope you'll join us next time for
more inspiring discussions
and expert advice.
Welcome
to IT Intelligence,
where we explore the latest trends and ways
to leverage artificial intelligence
in the IT industry.
Learn from innovators and IT leaders
and reveal the opportunities and risks of the
AI era.
And now,
(00:23):
here is your host,
Barnes Lam.
Hello. Today's
guest is Thomas Elliott,
a disruptive tech leader and trailblazer in artificial
intelligence,
cloud native
systems and scalable innovation.
With over two decades of experience in leading
(00:44):
startups
through exclusive growth and major acquisitions,
Thomas has become the go to expert in
building resilient,
high performing
engineering teams
while embracing the bleeding edge of technology.
From self heating networks to real time fault
detection,
(01:04):
he has been shaping the future
of tech before it had a name. He
is currently advising global AI and data strategy
as principal
at DataNash,
while also fostering executive tech committees
through Filio.
If you are curious about the real world
impact of AI,
(01:25):
team growing strategies,
what are coming beyond 02/2025?
This episode is one you do not want
to miss.
Welcome, Thomas. Thank you for joining the show
today. How are you doing? Thank you. Quite
the introduction. I'm doing quite well. Thank you.
Thank you again.
Let's start. You have led teams through massive
(01:46):
scaling and multiple acquisitions.
How has artificial intelligence
influenced your leadership and team strategy? Oh. AI
has been along for the ride for a
long time for me. And
a lot of people
see AI and equate it to generative AI.
My AI journey
started with products that we were bringing the
(02:08):
market back in 02/2004
at a security company called Entrust,
where we introduced
fuzzy logic mail filtering,
and then later real time fraud detection. So
how has it influenced my
perspective on leadership? That's a really, really good
question because it's meant that my focus
has been on driving results through innovative tooling.
(02:31):
I've always wanted to make sure that the
tools that my team had access to were
as innovative
as the tools that we were pitching and
bringing to market. So being a leader of
service teams, of field services teams,
where
we're selling this innovation
and
we are pitching this vision of the future,
(02:52):
being able to
keep up with our product and being able
to introduce the collaboration
and
the capabilities
of what makes high functioning teams
work at that next level
has always been
something that has been meaningful to me in
terms of the technologies that I would gravitate
to. Now generative AI is really
(03:15):
opening the bounds of accelerating
the capabilities
of individuals.
And we're just the beginning stage of, you
know, of being able to understand what this
means
from a leadership perspective
of what new shapes our teams are going
to take.
Great. Thank you. This is really good for
(03:35):
Satway for our next question.
Obviously, you just mentioned, alluded to earlier, you
have been working with AI since early two
thousands.
What excites you the most about AI in
02/2025
and the decades ahead? You mentioned about generative
AI and all these different things that we
just started. Can you give to our audience
a taste of what you see from your
perspective? There's a great quote that I hear
(03:57):
repeated frequently, and that is, when you're working
with this technology right now, every day you
wake up, I am working with the stupidest
version of this technology
that I will ever work with. Interesting.
So
despite
the capability
and the acceleration
potential
that's baked in today,
(04:18):
This is really the least capable
technology
that I'm gonna work with in my lifetime.
Interesting. Wow. And it I guess this comes
from a telecom notebook where I have, like,
a ten year perspective
of what to expect and and
where
skate to where the puck is going. Mhmm.
Yes.
A fellow Canadian. Yes.
(04:39):
The great Canadian.
And that is
it really becomes
there's an element of cognitive incoherence
Mhmm. Of being able to understand
a future self
and what that reality is. The only thing
that we can do is look at patterns
that are evolving now
and try to predict
(05:00):
the patterns that will spin
on that trajectory.
So
we think of we think of
the improvements
in cognitive capabilities of AI. Right now, working
with transformers,
we're
seeing artificial intelligence
that is able to
(05:20):
begin
faking out reasoning,
where it's a large language model and it's
trying to
predict
a conversation
or what makes sense
in terms of a response. And we all
know that AI can get ahead of itself
or
in the absence of facts and its model,
(05:40):
completely hallucinate. I like that one.
But it's in the future
with how models
and
new versions of intelligence will evolve,
we'll be able to
get beyond just the frontal cortex
of the brain, which is just a reasoning
engine, and be able to
(06:01):
look at additional theories of mind where we're
looking at short term,
long term memory, how that interplay,
how that ends up being self referential
in being able to evaluate
responses.
And then, of course,
implementation of doubt.
Doubt is a highly
functional,
(06:22):
self referential cognitive model
that AI doesn't have
yet.
And it's part of what makes
a model
of what would make a model
questioned its response before speaking.
Wow. That's something that we've evolved ourselves.
Right. So
this is the thing that blows your mind
(06:43):
is that even if we get to the
point where we have been able to orchestrate
that or or program this and create a
model for it, there's very little in terms
of observability
that will allow us to understand
why it's making that decision. It will be
not purposely obfuscated
from us. It will be part of a
(07:03):
thought model that is
core to the intelligence itself. It's kinda like,
why would I ask Barnes?
How did he come up with that? And
there's, like, a multi factored response. And ultimately,
it's just it's this response felt right. But
when you say felt right,
so there's a certain part
of your
decision making
(07:24):
rests on human
rather than rests on artificial intelligence. Currently,
absolutely, yes. The decision making
is assisted by artificial intelligence. Right? Okay. Right.
Right. It's about having another member of your
team. Right? Mhmm. And right now, that team
member is at a point in its, I
(07:45):
guess, cognitive capacity. Mhmm. That's somewhat limited.
Mhmm. Right?
It can
summarize
really, really well. It can even
do some fantastic
correlation.
Mhmm. But you give it a lot of
facts. You give it a lot of interconnected
stories. And it has
well, let's be honest.
(08:06):
It has very similar
challenges
that a wet neural network that our brain
has. Mhmm. In that, it can get a
little mixed up.
Right. Connections.
It's a concept of entropy,
where you have connections being made where they
should not be made, and then you have
connections that are being made where they maybe
were not needed.
(08:27):
Right.
I understand. Understand. That's just really, really good.
I never thought that way at all.
This is I think my audience would love
to hear from you about these questions, I
think.
What do you see the biggest buying spots
companies still have when it come to implementing
AI at scale?
(08:47):
Pick one.
You know, where it's Yeah.
There is
I think that
moving
into
there's a long tail with agency right now.
And everybody's hot for AI agents.
We're so very early
in the beginning of that. But it's where
(09:09):
you end up with multiple
Think of it as like microservices.
Right. Right? Mhmm. When we launched microservices,
not everybody really created a microservice very well.
We're now in the stage of AI where
people are building an agent
and
sometimes
it's this could have been solved with a
(09:30):
regular expression.
Right? Yes. Or it could have been solved
with a bash script. Mhmm. That compute would
be a hell of a lot cheaper. Right.
But when you really branch into
the elements of agency of being able to
do correlation of data Mhmm. And
(09:50):
be able to
have,
user profiling
that's connected with a correlation engine.
I think that is
going to be one of the most valuable
elements
of what telecom companies can use of what
basically any customer
that has, you know, hundreds of thousands of
(10:11):
of customers can do. And they want to
be able to understand better, how is it
that customers are using my software?
What are their challenges with using my software?
And what is my software missing that my
customers are needing?
And AI can be an element in helping
us
understand what that is, prioritizing
(10:33):
what needs to be delivered,
and helping product managers
build the future of what it is that
they are building as a product,
as a collaborator,
not as a replacement for what it is
they already have.
As an example,
one company that I
see great things for is
what Comcast is doing with their
(10:55):
Xfinity product.
What you what is singularly unique about what
Comcast that Comcast has done
is that they are pulling together this product
that's not just a network in a box,
but
a full
customer profile
graph management system.
(11:16):
Mhmm. And
that profile
of their customers'
product use engagement,
family demarcations,
and
who has access to what and when.
And all of that
forms
an incredibly valuable knowledge graph Mhmm. That will
(11:36):
eventually be
incredibly useful for artificial intelligence.
Anybody who's building knowledge graph of their
of their customer activity
and
of how it is their product is is
being used Mhmm. Will have a leg up
on
the next generation of software that's being built.
(11:57):
Awesome. That's really the important point. Thank you.
You have worked with
everything from real time fault detection to self
healing networks.
Which use case do you think will define
AI's
in mainstream
moment in enterprise tech? You know, from coming
from telecom,
(12:17):
I think it's a really great
space to have been exposed to AI
Because telecom enterprises have been running AI innovation
since the seventies and eighties.
Right? If we go back to DMS
and early electronic switching,
(12:38):
there was
the early stages
of creating
rich data sources
for
automating heuristics
and
doing
predictive fault or anticipating
network faults.
And everything that's been happening since then has
been rolling forward
(12:59):
ease
easing in tools of
automation. We went through phases of monolithic software
of of
where you had bounded
hardware and software being shipped. And then we
went to
decoupled hardware and software with monolithic
software elements. And then we went into cloud
(13:20):
native software where we
decoupled the software and introduced microservices
that are interoperating.
All of these patterns
move us in the direction
of
greater senses of scale,
greater senses of programmability
of services that we're creating.
So that the elements that I see where
(13:41):
AI comes to play
is not that AI ends up being an
overriding
or super controlling element,
but
as a application collaborator.
And
that's where we have
efforts that are going on right now,
where we're introducing
(14:02):
common API frameworks.
In telecom, we have efforts like
Aduna and,
Kamara,
which is the creation of
open
intelligent APIs.
And these are APIs that you could have
applications interact with. You could have AI agents
(14:23):
interact with. So it's about
building out on those patterns.
You're
leaning into openness
where
you're allowing
communities of developers
to participate
with the interfaces that you're developing.
And then also
embracing open source technologies.
(14:44):
And all of this intersects with AI because
if you build good interfaces, if you build
good scalability, there's nothing that prevents
AI from becoming a consumer of those interfaces
and interoperating
with those interfaces.
So that those applications don't just empower other
applications,
but they empower
(15:05):
additional intelligences.
Mhmm. That's a key part. That's absolutely. Thank
you. You have helped companies
reduce delivery complexity by up to 60%.
How much of that was due to AI
versus leadership and process?
Oh, man.
You really hit the nail on the button
there because you don't get to one without
(15:26):
the other. Talking about processes brings me back
to the trauma and scar tissue from, like,
ISO
certification.
So I tend to avoid using processes
and I turn to lean into ways of
working.
It may mean the same thing, but I
just feel ways of working is just a
better way of describing what we're getting to.
(15:46):
And that is, if you establish good patterns
for being able to work together as a
team and to collaborate with other teams,
and those
ways of working are predictable,
documentable,
and can operate at scale,
then you have the capability of being able
to add in
(16:06):
artificial intelligence
as a collaborator and accelerator.
If you don't have that, I'll just lean
into it. It's garbage in, garbage out. All
the only thing that AI will do in
that case
is put you on a turnpike to failure.
Oh, direct. We will do. You may get
to a failure point very, very quickly. Quickly.
(16:27):
Faster. Right. Right.
You are a founding member of AI Circle,
and I really encourage our audience to check
it out. What trends are emerging in that
community
that others should know about? The AI circle
community is it's been a really great organization
for people that have this keen interest in
(16:47):
advancing
artificial intelligence
and want to be able to have a
community
where you have like minded people
that are able to, you know, throw spaghetti
on the wall. Mhmm. I like that.
Right? Yeah. Yeah. It's messy. It's insightful.
It's camaraderie.
And most importantly,
(17:08):
it creates a sense of community
of people that are speaking the same language.
Mhmm.
I can't really point to one single thing
that's coming out as
as a trend, but I'd say the common
focus
of people in AI circle is looking at
practical and predictable uses
(17:28):
of artificial intelligence
that will stand up in production,
that are able to demonstrate the best that
the technology
can bring to a solution, and also be
able
to
show how it
accelerates
our capabilities
as human beings.
Well, so Because it's
(17:48):
the one thing that's clear in this community
is that nobody is looking at these solutions
as
a replacement for human effort.
Mhmm.
It's all about
the acceleration
of our efforts.
Perfect. Thank you. With your global
work from Poland to Brazil to US,
(18:09):
how do you see AI adoption
differing across regions
or cultures
in a general term?
You're pulling the hard ones.
It's
I feel somewhat disconnected
over the past
six or so months,
having just started recently working with
global teams again. But I think one of
(18:31):
the things that's universal
is
that no matter where it is that you
go,
if you're connecting to people elsewhere in the
world,
they all share the same concerns over AI.
There is a very real concern of,
will this replace me? And
will this
efficiency result in
(18:52):
reductions?
My sense of optimism
is all around
the empowerment of individuals
to be able to do more, to be
able to understand more, and to be able
to move their efforts
forward
with a greater understanding of why
they're doing those things,
of how they're going to be doing those
(19:12):
things.
And
I think that
if we expand
AI's influence
beyond just
large language models and
responses out of GPTs.
We see the real
capabilities
being
going back to the the
some of the very early origins
(19:33):
of AI, which is natural language processing.
And you had
IVRs
explore this back in the eighties
with voice activated,
menuing systems
on IVRs.
Right. And now we have
universal translators on our iPhones.
Mhmm. Where I can go to Japan or
(19:55):
Thailand or Russia
and be able to point my phone at
a at a sign. And it renders
a foreign alphabet
into
decipherable English.
Mhmm. Or at least usable English. Yes. Usable.
That's right. I think any kind of technology
like that makes us feel less like aliens
in our own world.
(20:16):
Absolutely.
Thank you. I promise, I mean, I can
go on for another hour, but I don't
I don't think we have that. But, you
know, I just I already like to my
audience, I have questions. I already go through
half of my questions.
But anyway,
let's go on. What advice would you give
to a tech leaders
preparing for AI driven future,
(20:36):
especially though just now starting to embrace it?
My advice is
when you're creating applications
or when you are looking at your enterprise,
look first as to
what you're actually doing.
And
do not be afraid to answer the question
of why are we doing this?
(20:57):
I think one of the key patterns
of embracing
an AI first way of thinking
is
one of looking at a
process that begins
to simplify
complexity.
Because once you are able to understand what
you're doing
and
(21:18):
you can put
a workflow or documentation or repeatable success around
ways of working,
That's where you can introduce these accelerators.
If we follow the patterns of what's happened
in telecom,
you've
moved from these
complex
service orchestrations
and complex
(21:39):
delivery
of AI
and increasing use of AI and all of
these APIs throughout the network. We're now at
a point where
I'm not saying the network at this point
is self aware,
but it finally has enough self respect to
clean up after itself.
And that is really
one of the key elements of what AI
(22:00):
enables.
Cool. I promise. Not too much longer. I
got a couple more questions and and that's
it.
Depending on your career, what do you consider
your most significant achievement?
And what lessons did you learn from it
that you continues to apply today? You know,
I go back and think about my first
trip into leadership. I was
(22:21):
beyond
fortunate
in that I was a technical resource that
operated the peak of the organization,
top level contributor, the kind of person that
nobody wants to become a manager because they're
gonna lose their top contributor.
Interesting. Yes. I agree. I made that jump
and
I had a leadership team that did not
(22:43):
make it easy for me.
There was a lot of scar tissue that
was developed in that transition
and they held me
to a very, very high standard.
Everything that I learned
Well, not everything, but a lot of the
foundations of what it is that I learned
about leadership, about people, about
(23:04):
leading others,
and making sure that others can succeed.
I learned from
the people that took me under their wing
and
explained the importance
of being able to
not just
manage people,
but truly
become
(23:24):
a valued part of
the careers of others.
Being invested
in where people are taking their career,
how they're contributing as an individual.
A sudden realization is that a lot of
the core fundamentals of what it was that
I've learned there
go back to establishing some of the core
(23:44):
fundamentals of what is is I expect of
artificial intelligence.
How can I develop this intelligence
to be better than it is today? That's
a really good
point that you just made.
Thank you. Reflecting on your career,
what do you consider your most significant achievement
and
(24:05):
from your perspective
that you can apply today? At my Canadian
start up, Scenics,
there was a period
of time there
where
a new leader was introduced into the organization.
And
I had
what, at the time, I had perceived as
a personal setback.
And it basically felt like I
(24:28):
was taking a back seat,
a backward,
regression in my career. And over a period
of six months,
I really
leaned into that
and used it as a pivot in my
career
to become
highly specialized in distributed databases and cloud native
ways of working. Mhmm.
(24:48):
Ultimately, that level of technical expertise,
it steered the company in the direction
that it needed to be. Because
as much as I went in there and
did a reset of my career,
I also ended up delivering a needed reset
when a pivot
to the entire organization
(25:08):
because it ended up changing
the patterns and how it is that we
delivered software,
fully embracing cloud native automation.
It reduced
the level of effort for deploying our software
from, like,
weeks
to
a couple of hours if you count if
if you count downloading all of the container
images. Mhmm.
(25:29):
And then patching was a trivial exercise that
previously was, like, a full day's effort. We
reduced it to, like, five minutes. Wow.
So the infrastructure that we built up to
lead us to that direction to improve our
capabilities of delivery
ultimately ended up being the differentiator
and a reason
(25:49):
for acquisition
because we were able to demonstrate
a at scale
cloud native
product operating at a tier one carrier
that was
averse to change.
And
we were the one of the first cloud
native apps to operate
at this carrier,
and they were willing to go to bat
(26:11):
in a public press release
to talk about
how our product and our ways of delivering
the product
is changing
the way in which they're operating data centers.
So overall, I'd I'd say that that would
probably be one of the most significant
contributions that I've made in my career
of
being able to take something that looked like
(26:32):
a career regression
and turning it into
a success for the broader organization and then
carrying those patterns on into Ericsson who acquired
us. Mhmm.
Wow.
Amazing. I mean, obviously, you you come a
long way.
And, you know, your journey so far,
I'm sure that any successful person like yourself,
(26:54):
expertise like yourself, there are some challenging moment,
some hurdles that you have to overcome
to succeed in who you are today. What
are those hurdles?
Getting out of the way of my own
ego. I love it.
There is,
when you're going through a transformative
period of your career,
there is
(27:15):
an element, you know, I don't want to
get too hippie on you, but
there is an element of ego death
where you are surrendering
some of the ideas
of the foundations of what it is you're
believing in. And
whether or not functionally,
they can play a part in
(27:37):
where you need to be able to take
yourself.
Being in a transformative
period of your career means that you have
to be able to be open
to
letting go
of that
and to embrace new patterns and new new
ways
new ways of working, new ways of thinking.
Mhmm.
And it's not to do so just because
(27:57):
it's the new way of thinking.
It's because
you're either embracing patterns that are keeping you
in place or you're embracing patterns that are
moving you forward. I like that one.
Never put it that way. I like it.
I really like it.
Thank you.
I like it a lot, actually. What are
three tips that make you successful in what
(28:18):
you do?
I have when I'm interested in something, I'm
really, really interested at it. And I dive
deep into it. And
when you're in a leadership position and you
can understand something deeply,
it doesn't mean that you
are necessarily going to get your hands dirty
and start coding. But you're going to have
(28:38):
a better appreciation for what the people that
you are leading
are doing. Mhmm. And you can look at
their patterns of working with greater clarity.
And
you can see,
sometimes,
another
or better way or streamlined way of reducing
(28:59):
complexity,
of simplifying ways of working
with a full appreciation of what it means
to be working with the technology
at hand. Mhmm.
Wow.
I now I understand that
working with artificial intelligence really changing the way
you look at things so so deep differently.
That's amazing.
You accomplished so much in what
(29:21):
in your career. What are your goals and
aspiration in 02/2025
and beyond?
So, you know, with
my some odd years
behind me,
I'm now at the point where I am
looking to be able to get into advisory
and consulting roles, where I'm going to be
(29:42):
able to take the things that I've learned
and the patterns
that
have served me
and
my vision of
how things could be working. And I wanna
be able to begin sharing that process, that
way of working with others
so that I can
help them enable
their teams,
their styles of leadership
(30:03):
on
being able to
look at alternative trajectories,
look at
adding
continuous improvement loops into their workflow,
being able to
look at
how it is that their that their teams
can
be served better by a different style of
(30:24):
leadership or
levels of focus to their organization.
Cool. Last question.
If you could give yourself some tips for
your younger
self,
what would that be? Be less afraid of
letting go. Be less afraid of letting go.
Okay.
Be less afraid of letting go. There's a
difference between
(30:44):
being proud of your accomplishments
and
being able to
believe
in
certain patterns.
But there comes a time where
sometimes those patterns may not
be as helpful as they were.
Mhmm. And you need to be able to
let go and embrace these new patterns,
(31:06):
these new ways of working. And that applies
personally,
as much as it applies to
patterns in
organizations
that may have grown out of necessity,
but now
are collapsing under the weight
of legacy.
So that openness,
that lack of fear
(31:27):
to be able to embrace new patterns is
so it's so transformational,
and it's so
it's become so much
behind,
I guess,
my own personal operating system.
I like that. I like that one. Yes.
Thank you, Thomas. It has been wonderful
to having you on the show.
(31:47):
That's a wrap on today's conversation with Thomas
Elliott,
someone who truly
exemplifies what it means to lead with both
vision
and execution in the AI era. Whether it's
building teams that scale with soul, redefining what's
possible in cloud native infrastructure,
or advising the next wave of AI leaders,
(32:08):
Thomas is a force
at the intersection of technology
and leadership.
If you enjoyed this episode, please like, follow,
and share with someone thinking about how AI
will shape the next decade.
And don't forget to connect with Thomas on
or
at a website, which I will be
listed.
(32:29):
Until next time, keep learning, keep building, keep
pushing boundaries.
See you next time on IT Intelligent Podcast.
Goodbye for now.
Thanks for joining us for IT Intelligence.
If you enjoyed this episode,
subscribe and leave us a review so we
can reach more tech heads and industry thought
(32:50):
leaders
like you.
For more information,
extended show notes, and additional resources on today's
topic, visit ITIntelligence.co.
We hope you'll join us next time for
more inspiring discussions
and expert advice.
Popular Podcasts
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
24/7 News: The Latest
The latest news in 4 minutes updated every hour, every day.
Therapy Gecko
An unlicensed lizard psychologist travels the universe talking to strangers about absolutely nothing. TO CALL THE GECKO: follow me on https://www.twitch.tv/lyleforever to get a notification for when I am taking calls. I am usually live Mondays, Wednesdays, and Fridays but lately a lot of other times too. I am a gecko.