Dan: okay, and welcome to the next
episode of the Quantum Divide.

I'm on my own again this week, but
I'm very happy to be joined by...

Juan Moreno from AWS.

I'm really looking forward
to this conversation.

We've been building up to this
for a while, and I think we've got

some great topics on the table.

Juan, you're a, a developer evangelist
or a developer advocate for the

quantum networking team at AWS.

let's start like we normally do.

Give us a bit of an overview, who you
are, what your role is, and we'll go

Juan: from there.

Sure thing.

Yeah both terms are okay.

Developer advocate or
technology evangelist.

Depending on who I talk
to, I use one or the other.

And I've been with the quantum
technologies team in AWS for the past

three years, both in quantum computing
first and lately in the last year or so

in the center for quantum networking,
that is more focused on networks.

But basically I've been over 10 years
with AWS overall, it's been half of

my career in this company so far.

Dan: Fantastic.

So how did you get into quantum?

Juan: Yeah.

That was, that's a bit of
an unconventional path for

me, I'm curious in nature.

I like to read and to learn new things.

And in August 2020 AWS
launched Amazon Braket.

That is the service for which we
give access to Various quantum

computers to customers with a
single development environment.

So they can test them and so on.

And at the time, I was just basically
checking around this new service.

What it is about.

I didn't know much about quantum
computing or quantum mechanics whatsoever.

And I started to get involved
with it and learn it.

And it was a good feat in the sense
that as you very well known, this

is a very academic in In at, at
the heart of it, at the core of

it is very academic environment.

So it fit well for me to come and
bridge that gap between the industry

and the academy and try to explain
these concepts to people that are

not necessarily familiar with quantum
mechanics and how quantum computing

and how quantum networking works.

So I started to co collaborate
first informally with the team

and then in a more formal way
in, in the role that I do today.

Dan: I think any kind of education
and communication around quantum

mechanics into an IT company is
probably one of the most difficult

jobs on the market right now.

Just because of the breadth and the
depth of understanding that's needed.

Do you have a, how did you tackle that?

I understand you've got a
background in physics, right?

But that's fairly recent.

Did you do that in order
to launch into this

Juan: role?

That's right.

It's been in parallel.

Actually, my background is in
something completely different.

So I realized very early that this
wouldn't be like, I don't know,

learning Kubernetes or networking
concepts, classical networking concepts.

I really need a good foundation
of linear algebra and

basically the theory behind it.

You can work without it.

But if you really want to
explain things to others.

In a way they can understand it's
better to build a good foundation.

So in this case I started to study
physics a couple of years ago, just to

be able to level up my conversation with
people that are obviously a lot more

knowledgeable than I am on this field
and be that, I don't know the voice of

trust or credibility that probably AWS
customers want when they are discussing

a such important project like this one.

Dan: Very good.

Hey, let's, you mentioned Braket,
let's go on to AWS now a little bit

and yeah everybody knows AWS, right?

Your cloud services, your pay
as you go cloud services, the

infrastructure, global infrastructure.

What about AWS getting into
quantum in general, what can you

say about that as a, as an intro?

Juan: Yeah, so it's a little
bit similar to a similar story.

I would happen with AWS.

First people used to say
Amazon is just selling books.

And now you guys are getting
into the cloud and renting space

for people to do computation.

What is all that about?

It's a similar journey in the
sense that the easiest answer

that I can give you is that.

Customers are really asking for it.

So that's why we want
to go into this space.

There are certain um, advancements
and interesting fields of study in

quantum computing that we're starting
to put together the 1st prototypes and

very R and D based quantum computers.

And we wanted to be part of it
and especially see how this could

be useful for our customers.

We build Amazon Braket as a way to try to
democratize the access to these computers

and lower the entry barrier to, for
people to use them, um, and change the

typical payment method or payment model
that is used in the industry to something

more AWS like in the pay as you go.

So customers can try different
types of computers and.

And decide whether they, which one
is better for them, according to

their experiments and try different
things in a pay as you go model.

But since this is a very academic
kind of environment, we also

started to participate actively in
building our own quantum computer.

We have a team at the Center
for Quantum Computing.

Based in Pasadena, that is actually
working in a superconducting loops

type of quantum computer to try to
add that to the community and our own

effort on putting a computer out there.

That's besides Braket and probably
the other part that is important

is the quantum solutions lab that
is our professional services team.

And those guys, what they do is
basically work with a problem of the

customer backwards, trying to build
POCs that are quantum computing related.

Sometimes they are also HPC related or
even machine learning related problems,

depending on the kind of goal that the
customer want to achieve, but they want

to get the customers quantum ready.

In terms of quantum computing, by the time
these computers are ready and they focused

on that, um, and my favorite group is the
one that I work right now, the quantum

networks and the Center for quantum
networking started over a year ago.

What we do is basically investigate
the potential integration of quantum

networks in real environments
and integration in the cloud.

When we talk about short span networks,
but also we have an R and D team similar

to the center for quantum computing,
but for networking that are working

in prototypes and doing R and D work
on longer networks that span yeah.

Bigger distances.

That's in a

nutshell.

Dan: Yeah, great.

Yeah, lots of things to dive into there.

I guess.

First of all, look quantum
computer superconducting computer.

I didn't know that , you guys
were developing something there.

So that's pretty exciting.

Yeah,

Juan: it's pretty cool.

It's very interesting technology.

I'd say I'm hoping that one day we'll
be able, we will be able to, to yeah.

Put it out there.

The blueprint is there already.

This is 100 pages paper that explains
exactly what's our idea for it.

And it's available for people to see.

And maybe one day we'll
be part of Braket as well.

Dan: All right.

Very good.

Yeah.

So with Braket, I understand.

You're leveraging the best of the cloud,
you're providing what I would call,

the middleware between the user, the
developer, and the quantum computer,

you're democratizing the access to it, um,
but what does Braket do on top of that?

Does it have its own, Coding language.

I believe it doesn't it?

Is there some

Juan: right in order to make
it easy for people to use?

We don't want them to learn a different
language for each computer, right?

That is sometimes one of the
things that is stop adoption.

I have to learn a whole new language if
I work with this vendor or that vendor.

So what we do is to unify
everything In the Amazon Braket

language, that is Python based.

And basically, if you are dealing, for
example, with with circuit based or

universal gates type of quantum computer,
you basically can change your end point.

And see how the other device
behaves for the same circuit without

having to change anything else.

And that's handy.

And on top of that, we
offer simulators as well.

So you can tune the noise.

If you want adding noise or remove noise,
you have various types of simulations.

You can do before you jump into the
actual real device and all that.

As I said, it's with the same code as
long as the technology supports it.

Dan: Got it.

Super.

Hey, I'm sure we could talk
about Braket for the whole call.

But I know that we're here
to talk about networking.

So let's get into that.

I know we've got three or four
topics that we wanted to cover.

Let's start with the first
one, what is your team doing

when it comes to, , networking?

Quantum state over a network,
what, why is that important to you?

And why are you looking at it?

And what have you got in the skunk work
so to speak that you can talk about?

Juan: Yeah.

It's a very rich.

Interesting topic to answer the 1st
question, why we are getting into

quantum networks and investigating them.

It's because we believe there are
more secure ways of communication

looking into the future on what is
coming in the technology and how the

existing encryption mechanisms might
be under under the risk of being

threatened by the new quantum computers
that, Hopefully, 1 day will exist.

And we have a dedicated team in AWS,
the cryptography team that is working

very hard on post quantum cryptography
type of algorithms that basically use

more complex mathematical problems
to make sure that we create a cipher

suites that are synchronous There are
quantum safe as they are usually called.

But the focus of my team is on the,
in the hardware layer on trying to use

the properties of quantum mechanics to
ensure that the link is protected in a

way that do not depend on computational
complexity of an attack, but just taking

advantage of the properties of quantum
mechanics and the properties of light.

So on that line, we have Two main
teams, one of them is working

on quantum key distribution.

That is where most of my focus is.

And we basically want to test the existing
commercial off the shelf boxes that exist

today in the market that people can buy
and try and see how do they perform?

How do they compare to each other?

And how could they potentially
get integrated into the cloud?

And.

That's an exciting line of work.

tHe rest of the team is the purely
hardcore R and D team that are working in

networks are more advanced, more complex,
but also more difficult to implement.

We have a team based in Boston and we have
an agreement, a collaboration agreement

with Harvard University to work and
In, in, in building a quantum repeater

or a quantum memory that is basically
going to allow to, to make long distance

communications with this possible.

And that's the other part of
the, of what my team does in the

center for quantum networking.

Dan: Yeah, that's a lot going on.

So let's start with the
other part of the team.

First of all, that you mentioned doing
the hardcore um,, networking, as you put

it I guess the purpose of that is to.

eventually enable customers to get to
the cloud over longer distances, perhaps.

But what is it they're doing with
the repeater, that they're developing

something from scratch or are
they putting something together?

Yeah, it's,

Juan: It's devised based on the
properties of silicon vacancies on

diamond which basically allow to retain
The state of the photon long enough

to be able To extend the distribution
of this estate to a longer distance or

at least duplicate at the very least.

What QK is capable by itself.

QK is a point to point type
of protocol and technology.

And what we are trying to do is to
extend that, um, it's a challenging task.

Obviously there are.

Making good progress and 1 of
the requirements is that it still

requires very low temperature.

So same as in with quantum computers.

For that reason Being able to create such
a repeater in a very uh, repeatable way.

, if I can use the word like that, or
a scalable way is a challenge because

basically you cannot put a dilution
refrigerator in a, every a hundred

kilometers or so, or in a submarine
cable or things like that today.

That's the main challenge.

So the main focus, one of the focus
of our investigation is to actually

try to make these things more.

More portable and the feature deployable
to create that the quantum internet, if we

want to call it that way that we all want.

Dan: You mentioned silicon vacancy.

The only other type of diamond
modality that I've heard

about is a nitrogen vacancy.

Do you know that?

Obviously there's a
different element in there.

And I think the nitrogen
vacancy ones don't need cooling

down with a refrigerator.

Is that, do you know the
differences between the two?

Juan: No, I only know that it's
related to scalability and the

choice of using silicon over nitrogen
when we want to do many of them.

But I don't know, not intimately familiar
with the differences or other challenges

on actually building the technology
with one or the other, unfortunately.

Dan: Every time I have a conversation
with somebody, I end up with multiple

things I need to study afterwards.

I think that's what I'm taking away today.

hEy, let's move on to your
domain in terms of QKD.

What's the overall strategy there?

And obviously you're a very
customer centric organization.

So I imagine you start from the customer.

work your way in.

So this must be driven towards
connectivity to the cloud.

Is that right?

Juan: That's right.

Yeah.

We want to see, we're in
gathering data mode at this point.

We want to see what customers want to do
with this listening to the different use

cases and how are they going to use it.

And we want to start to To help
the people who know about this to

interact with these customers and see
what are the issues they run into?

What are the challenges of
the implementation and try to.

As much as possible for it, because
that's going to allow us in the future

to define our strategy in terms of
do we want to offer QKD within AWS

natively, or we want to continue building
this network of customers doing this.

It's important for us
at the moment to know.

How it's being used and that's
the main part of my work.

And while I'm starting to work
with various partners that

are helping us to get there.

Dan: So I'm picturing a lab with loads
of different QKD devices, um, and lots

of simulation and testing equipment.

And you might not be too

Juan: far wrong on that.

Yeah.

You might not be too wrong on that.

We want to see how they operate and so on.

yeAh, we are definitely testing them.

buT the most important, we could be
doing things in a, inside a room in a

lab all day long and that's definitely
useful, but there is nothing like the.

The customer feedback to learn about
really the implementation of a product

in real life and see how that integration
in the cloud, especially the integration

with the rest of AWS services will
give us the most valuable information.

And that's what I focus
most, most of my time.

I'm talking to those customers and
the partners that are helping us.

Dan: Okay, so look, I know that I
know a little bit about the AWS cloud.

architecture and, the fact you've
got data centers regions all over

the globe, and people need to
connect to those regions somehow if

they want to have a dedicated link.

You mentioned earlier on QKD is
a point to point technology.

And what I wanted to get to was how
do those two things Link together.

I would guess if you were thinking about
offering a QKD protected service at some

point in the future, then the customer is
going to have to host the opposite end of

their QKD link in a your data center or in
the data center where you're your cloud is

hosted or nearby or, whatever it is, the
right architecture, but it's going beyond

needing just an IP connection, right?

You need to have that quantum state
transferred across the network and then to

be able to calculate the keys and so on.

Could you elaborate

Juan: on that?

Yeah.

Yeah.

That's the most obvious use case, right?

If you want to bring this technology close
to the cloud taking into consideration

that is basically a layer two is very
physical level technology, you need to

start operating in, in the data center.

The direct connect service of AWS
is probably the most obvious choice

here, because that allows us to
get very close to our customers in

a typical direct connect facility.

Or co location facility, there
will be a, an AWS rack with an AWS

router and maybe two rows down the
street and that data center will be

a customer or partner router that is
hosting a QKD link all the way to the

customer offices or whatever data they
want to protect on their premises.

Right?

And What we are trying to do in with
a couple of partners, and the most

obvious solution is to basically
link those 2 things together.

Obviously, we do not support
in AWS QKD natively today.

There is no a series of QKD boxes that
can be linked together all the way

to the cloud, but there are solutions
that we can put together in order

to offer a quantum resistant end to
end link or connection if we can.

If we use a post quantum
cryptography, right?

The protocols that I was using before.

1st of all, to talk about the QKD
part only a customer might have

already several buildings or several
offices connected with these links.

As I said, they are point to point.

So there's an option to link them together
using what it's called a trusted node

configuration and that allows you to.

To protect extend the, dis the distance
as well and go all the way to the data

center . . And from there we, we have
the choice or the customer has the choice

to, to get a an standard cross connect
connection with direct connect and connect

to an AWS region and protect that part
of the connection with post quantum

cryptography, with with this quantum
safe protocols that way we are generating

keys on, on one side that are QKD keys
and transporting them all the way.

To the cloud using a PQC.

So they get the protection end to end.

This is not something that it's out
and available today, but it's an

interesting area that we are investigating
with our partners and customers.

We hope to be able to get it done
soon as a first step of integration.

Dan: Yeah, it's an interesting
hybrid model to try and connect

the end to end in a more secure.

Quantum safe way.

I had somebody earlier on an anecdote
about people often seeing QKD and PQC.

as competing but actually the, when you
think about it, when you think about

the hybrid architectures there's going
to be, , scenarios like this, where

, they're complementary to each other.

And, um, in this case, the QKD has a
distance limitation and it has a media

limitation, but ultimately you want
to use the keys providing they are.

Renewing themselves frequently enough,
you want to be able to use them into the

rest in the rest of the cloud environment,

Juan: right?

Yeah, you can think about it in
the defense in depth terms as

another layer of protection, right?

Something that you can
add to PQC to get a.

An extra layer of security, and
obviously the state of the devices is not

something that allows to be implemented
in data centers all over the world.

And that easily.

They are still the
standards are being defined.

There's not a clear definition on
how these things should work, or at

least not something that every single
vendor is very clear on how to follow

or how to implement, there are some
conversations going on right now on

how those standards should look like.

But it still is some walk to some
path to walk there to get there.

Dan: Oh, for sure.

Especially when it comes
to the QKD hardware.

That's going to be very bespoke.

Juan: Across every vendor, right?

There's a lot of different
technologies, right?

And so we work with some of the vendors
that are using continuous variables.

Others use discrete variables.

So the way in which you treat the light
in order to create your key material, it

can be very different depending on the.

And the implementation of QKD,
but at the end of the day, what

you get is key material that is
completely quantum in nature.

That is what everyone wants.

Dan: Okay.

So the next topic is,
it's a follow on really.

I know that in the AWS cloud
architecture, you have plenty

of, I'll call them edge nodes.

I think it started with that
post as basically the on prem

cloud managed infrastructure.

And there are others of
different scale and scope.

Does this fit into the offering that
you're going to be building, you think

as a way to provide, off the shelf,
quantum safe Remote compute or on prem

Juan: compute.

Yeah, this is how they are.

Actually, we had a partnership with
Spectral that is a startup in Singapore.

Their main focus is in satellite QKD,
but they are also getting into the

terrestrial QKD and they have been the
1st partner working with us to offer

this solution using edge devices.

Which is nothing but a small version of
the cloud in a box that is ruggedized and

that you can go and take anywhere, right?

This is very useful for certain
environments where there's a rough

environment or where you don't have an
easy access to the cloud, but you still

want to get the same experience being
able to launch for those familiar with

AWS your EC2 instances and S3 buckets
And all those nice terms that cloud

developers want, and at the same time,
being able to use those devices as the end

point of a quantum protected link, right?

If we use QKD and we have a.

Point to point network protected
by QKD can use these devices to put

them at the ends on both ends of the
connection and being able to launch your

client server application or whatever.

It's it matters to you using [QKD]
is an easy way to start testing the

technology in a very safe environment.

And it's also an easy entry point.

That is what.

This probably the solution is the
main feature of it is that it's very

easy to order a snowball or a snow
cone or an outpost server, as you

were mentioning and start testing
the link that you build with QKD.

Dan: Yeah, it fits very well.

I see that

Juan: it's a bit simpler than going
to direct connect directly and having

a presence in the data center that
requires a little bit more, it's more

involved that type of work, but the
snowball or devices is much easier

to implement than the other one.

Dan: Yeah, I guess you're still
going to need the QKD endpoint in

the data center somewhere, right?

Do you know if you'll be hosting
those in the future or I guess you'll

have all the options on the table.

Juan: There's no plans for that.

As far as I know, it's at the
moment, as I said, we want to see

how it's being done by others.

And we are supporting the people who
does this, which in this business to see

how could be implemented with customers.

Dan: Okay.

So another train of thought I had was
around the , the keys that are generated.

Let's say there's a QKD link
for a particular customer.

The keys are generated and
handed into the AWS , cloud.

There's obviously key management software.

But when it comes to the rest of the
cloud infrastructure, and any of

your applications, they don't know
that it's a quantum generated key.

I guess it's it's just something
that's generated locally.

And then , it looks like any other key,

Juan: right?

That's a possible implementation.

Yeah.

You could basically connect two QKD boxes
together start producing key material,

and then you can feed that key material
into a network device a link encryptor.

It's usually called, but it has many
names depending on the layer they

operate, but let's call it a link
encryptor that support and understands

these quantum sources to create keys.

There are more and more vendors that
are starting to support this form of,

key material, and what you do then is
to create a VPN at that level or an

IP sec tunnel between the 2 devices.

Once you have that, it's
basically a matter of plugging

your endpoints on that device.

And then when you run your applications
at whatever OSI layer that you want,

you are starting to consume in the
keys that are provided to you by

the fact that you are in a VPN, but
the endpoints do not need to know.

that the source is a
quantum source, right?

And that makes it an integration with
the cloud very seamless, very natural.

The difficult part is to put
those hardware boxes the QKD

boxes in the right places.

Dan: Yeah, I guess because the key
management can then be owned by

the customers, so you maintain your
AWS cloud security model where...

Ultimately it's up to the customer
to manage their end to end security

and you guys own the infrastructure
underneath That's right.

That's still fits very well with that.

There's no need to change

Juan: that the key
management is important.

You might have even QKD
boxes from different vendors.

Obviously a pair is going to be
always from the same vendor, right?

Alice and Bob, they need to
be from the same manufacturer.

Otherwise this doesn't work.

But after that, you can
basically in theory link.

Other devices from other vendors and
do this trusted node configuration

that I was talking about before.

And if you have a good key management
software as an overlay on top of it,

you should be able to transfer keys
in a safe way from devices that are

not necessarily from the same vendor.

And that's a part that we
are not directly involved in.

At the moment, but
obviously very interested on

investigating and test as well.

With people who does this,,

Dan: Yeah.

So seeing as that you guys are in
the test and research phase still,

and just familiarizing yourself
with the different products on the

market, I wouldn't ask you to do any
comparison or anything, but if you got.

Any view on trends or perhaps
the common behaviors that you're

seeing among different QKD?

Systems things that you like
things that you don't like.

Juan: Yeah, as you said, it's still
way early and I will be doing a

unfair comparison for sure, because
I don't know all the details of it.

What I'm very clear about is that some
of these conversations are started to.

To be very like the mainstream, I'm
thinking on protocols like ETSI 014.

That is one that is being implemented by
almost every vendor that I've talked to.

And if I can write my letter
to Santa Claus on that, it will

be like every single vendor.

Complies to the letter with ETSI
014 is mandating because I will make

the future integrations much easier.

But that on itself is an unfair
statement for them, right?

For the hardware vendors,
because sometimes there are

things or running conditions.

That are not necessarily
defined very specifically.

So there is certain room for
implementation in the implementation

of the protocol that can vary
from one vendor to the other.

And that's still been.

being discussed and being defined.

Hopefully that's more or less
the answer to your question, but

probably it's, I didn't give you
exactly what you were asking for.

Dan: No, that's fine.

That's the authentication.

Juan: That's as political as I can go.

Dan: Yeah, I know.

I'm pushing you.

I appreciate that.

Are you checking things like key rate?

And looking

Juan: at the effectiveness and

Dan: yeah, and you
seeing any trends there?

Are they all in a good place in terms
of performance and, or is there a

big, are there two different areas
of device type or are they all?

fiTting in one particular
threshold, if you like,

Juan: I know there are pros and cons,
with each implementation of QKD, right?

I was mentioning before continuous
variables versus discrete variables, or

the key rates in some cases are higher
than in other cases, but the tolerance

to noise and using existing telecom.

Telco fibers is better in
one case than in the other.

So it's a trade off.

It's a little bit of the
same thing that happens with

quantum computer gates, right?

And sometimes the fidelity, it varies with
the execution time varies with the device.

Here is a little bit of the same.

You might choose a device that gives
you maybe shorter distances to, to

put two links together before you
start losing too much or seem too much

attenuation on your connection, but
however your key rate might be very good.

And you can see the
opposite as well, right?

You can go for a much longer distance,
but maybe your key rates are lower.

Dan: Got it.

Yeah.

Like a lot of quantum technology,
it's a very fine, delicate balance.

Maybe not so much in the networking
space as it is in computing, but

still you got to worry about.

Coherence times, you got to
worry about noise and so on,

Juan: right?

And there are many
implementations of it, right?

And we haven't tested them all.

There are still things that I still
like to get my hands dirty with.

I'm talking about continuous variables
and discrete variables, but there's

just 2 different classifications
on how to . Slice this cake of

getting implementations, right?

We have the measure device
independent option as well.

That adds an extra layer
of security by the way.

It's designed and there's not necessarily
the option to implement BB 84.

that is the most common protocol for QKD.

You can implement all the
protocols and there are some

areas that I haven't explored yet.

And I definitely interested on.

. I'm checking.

I'm pretty sure my team is as well to see
for our purposes, meaning AWS supporting

this one day or how customers are using
it for the workloads that we care.

What will be the best one of all of them?

Or what we will need to how do you say to
opt out in order to gain something better?

Dan: Yeah, I think you're in a strong
position as well to influence the

direction of how things develop
especially in standards uh, domain.

I know you guys are active in ETSI but
also as a key customer probably, or at

least partner to the QKD manufacturers,
then, uh, they'll, they need to

sit up and listen to, any kind of.

Outputs that you have from
this program in due course.

Juan: Yeah, you're right.

We have people in who's part of ETSI.

You are totally right on that.

Our approach here is to just bring our
data and not try to influence in one,

one way or another, because we want to
be part of the conversation and not to

impose a certain way of doing things.

We aim to bring our data, what we
are seeing with customers and that.

On itself should be useful to,
to define how the protocols

are going to look like, right?

That's the purpose.

Dan: I've got to ask
you about entanglement.

Are all of the systems you've looked at
entanglement based or a, most of them?

Prepare and measure because that's
the big difference in the way

that these systems are working

Juan: on paper.

Yeah, most of the the ones we work
with are prepare and measured as

the top uh, key, uh, quantum key
distribution boxes out there are

based on prepare and measure.

Dan: And that's basically
how BB84 works, doesn't it?

In terms of receiving the photons
and then making the calculations of

both ends based on the messaging.

Which ones you received and which
ones you didn't receive and so on.

Juan: That's correct.

Yeah.

If you go usually at 3 percent
of errors, you consider that as

an attack or an eavesdropping.

And then you consider that key material
is not usable and you start over again..

Dan: Okay, thank you Juan.

So a bit more about key management
and I'm jumping about a bit.

I've got a habit of doing that I know
that you guys are looking at something

around centralizing key management
You must you have a key management

application in the cloud already, right?

Are you looking at how?

Multiple QKD systems in a customer
can be managed and, telemetry

can be captured as well as key

Juan: management.

No, so integrating this
with our KMS system is today

completely out of scope, right?

The team is working on supporting, and
I believe they already support some

of the PQC classical Cypher suites
in KMS, uh, but we are not getting

into the business Of trying to be
offer key management for various

vendors, various hardware vendors, but.

What we do, and it's relatively
straightforward integration is

to extract the useful information
from the key management system

and the monitoring systems of
these devices into a visualization

layer that customers can see.

And it's very easy to
implement in the cloud.

we Have another partnership with
Lux Quanta, that is one of the

QKD vendors that are actually
working with us on this and.

They will very soon be offering
this solution to their customers.

Basically, it's a matter of bringing
all the key metrics of how your key is

used, key buffers, key rates, errors, and
all this useful stuff into a front end.

That customers can consume, it can be a
Kibana based database or Grafana, sorry,

Grafana based database or visualization
layer with time log series and a database

for logs as well to analyze those logs.

And it's this is very easy to
integrate in the cloud, because at

this point, number 1, there is no more
secret information that needs to be.

Protected it, this is the control plane.

So it's nothing that
nothing secret on itself.

A number 2 is very easy
to integrate in the cloud.

We have many years of experience of
building resilient applications that

scale and that provide that seamless.

Experience for the customer, they
will have a dashboard where you

can see everything that is going
on your system and then you can

analyze logs and see what's going on.

That's something that is more
classical in nature, but the

information comes from a quantum.

Dan: Yeah, of course, once you've
pulled the telemetry and then you

can do whatever you want with it.

And anything that you can do
to make I suppose the AWS cloud

more attractive people to, to
consume is what it's all about.

Juan: As you can see, all I'm trying
to do here is to go around and get as

close as I can of from the cloud as
the technology allows me to do today.

Of allows us to do today.

Of course.

Yeah.

And this is a ver a very nice way.

Yeah, it's a natural fit,

.
Dan: Excellent.

Thank you for all of that.

I guess just looking forward a
bit in terms of AWS strategy or

goals that you have personally with
this technology what's the next

six, 12, 24 months hold for you?

Do you think that you can
share with the audience?

Juan: Sure thing.

So it's in the name of my title, right?

Trying to be a technology advocate.

I want people to understand what are
the benefits of using this technology?

I want to keep explaining what
are the difference between PQC and

QKD and what benefits it bring.

If I'm placing myself or Envisioning
myself 1 year from now, I'd like to

have a lot of companies that have
come with us to the edge, either

in direct connect facility or using
the devices and they are testing it.

They are pushing the limits of the
technology, defining the standards

and we have a lot of valuable data for
us to decide what's going to happen

in 2 years, 3 years time from now.

How are we going to use this
technology and keep innovating on

the name of the customer, basically,

Dan: and in terms of getting closer
to the the quantum computing part

of AWS, does that fall more into
the repeater and quantum memories?

Development that the rest of the team are

Juan: doing that's a natural
use case for quantum repeaters.

Yeah entanglement distribution that, that
is based on that technology is a natural

fit for things like private computing and
things like that, that will ensure that.

That you can use quantum computers and
connect quantum computers with each

other in using also quantum technologies.

foR QKD is this also
potential integration, right?

But they are not different to what
we have discussed today, right?

Treating that quantum computer
just as another end point that is

going to be sending data somewhere.

And we have some interesting things there.

We are working on, but that.

Dan: Great.

Oh, that's great.

Maybe we can have you back in a year or
so to talk about some of these things.

I'D love that.

Me too.

Yeah, definitely interested to,
to hear some more about your

repeater in due course as well.

So listen, one final topic an open ended
one about academia and open source,

are there any projects that you're
tracking or involved in or perhaps.

that your colleagues are involved
in that you're particularly excited

about and would like to share.

I

Juan: don't know much in terms
of involvement, but things that

I particularly like, try to
read and I find fascinating.

It's all about for example, the
OpenQKD project that is aimed to create

and extend a secure communication
network based on in Europe.

That's a pretty, pretty cool project.

But also there are for
people to try at home.

Let's say there are tools like
the QKD sim that allows the

simulation of various QKD protocols.

And so that's something that people can.

Can go and check libraries that
are also out there the Q crypto is

a library in Python that is open
source and basically allows you to

understand the principles of QKD.

This has been super useful for me
to wrap my head around some of the

concepts in a very practical way.

In a more general term, I.

Particularly fascinated about satellite
is not something that we are heavily

involved today, but just because our
partnership with the spectral, I get to

learn a lot about it and there is a lot of
work out there on space based to ensure.

Global secure communications.

This is following the experiment that
the Chinese did with the satellite that

you, I'm pretty sure you know about.

Dan: Yeah.

I find that whole topic
really interesting.

It intersects with some of my work
at at Cisco, also in the LEO space.

And yeah, the combination of the two
technologies is pretty fascinating.

In fact, there's been lots of
satellites launched and tests

ongoing and plans in the pipe.

It's amazing that they can hold a single
photon source over that kind of distance.

Even with a static satellite
it's just it's just amazing.

And then when you think about the speeds
of traveling in LEO or beyond, and it's

just amazing that technology can work.

When it comes to entanglement distribution
and that's obviously a great place to

do it from because of the distances
you can get and with things like

phase stabilisation and stuff and
all the different error correction

that can be built into the forwarding
of Photons over free space optics and

you yeah, super interesting domain.

Juan: It's a matter of
infrastructure to mature, right?

And being able to implement
these things at the scale.

That's what we all want.

I think for this type of technology,

Dan: Hey, maybe we should do a
quantum satellite Discussion at

some point in the future as well.

Juan: I know a couple of people
in my team that would be very

pleased to chat about that.

It's fascinating to talk
to them, to be honest.

They are the real experts on this,
and you can see that sparkle on

their eyes when they talk about it.

And that's always very, I don't know,
to me, it's very refreshing to see

people that really what they do.

And they are experts on
their field, obviously.

Dan: Yeah, excellent.

Okay.

One thing I've got to mention because
you spoke about QKD simulator

or the kind of open QKD model.

Just you may not already know,
but my colleague, Steve, he

normally runs this podcast with me.

He's developed QNetLab, which is a
quantum network simulator fairly recently.

So you should check that out.

We're going to probably the next
episode or the one after that.

We're going to do a
session talking about that.

So keep your eyes

Juan: peeled and I wasn't aware.

Dan: Yeah.

Yeah.

And happy to have a followup call with you
on that, but anyway, we're going off topic

again, as I often do uh, Juan, thank you

very much.

Fantastic conversation and
covered some great ground there.

Let's let's reconnect.

At some point, but for
now I'll say goodbye.

Juan: Thanks very much, Dan.

Thanks for having me.

Dan: I'd like to take this moment to
thank you for listening to the podcast.

Quantum networking is such a broad domain
especially considering the breadth of

quantum physics and quantum computing all
as an undercurrent easily to get sucked

into So much is still in the research
realm which can make it really tough for

a curious IT guy to know where to start.

So hit subscribe or follow me on your
podcast platform and I'll do my best

to bring you more prevalent topics
in the world of quantum networking.

Spread the word.

It would really help us out.