Chris and Selina are joined today by Sven Przywarra, he started-up Live EO, wich uses artificial intelligence to analyze satellite data, they have recently generated over 1.5 million dollars in sales and have around 50 employees. In this episode they discuss data science and monitoring assets from space.
Broadcasting from the commodity capital of the world, Zurich, Switzerland, this is insider's guide to energy.
This addition to insider's guide to energy is brought to you by FedEx.
Go to www.fedex.
Dot com For more information.
Welcome to insiders guide to energy.
I'm your host, Chris Allison.
With me this week is Selena.
Selena, welcome to the program.
Hi, thank you so much.
I am so excited to have a new co-host because you are our data science expert and you are part of our new data science mini series.
So welcome to insiders guide.
This is our.
Audience first chance.
To hear you.
Thank you so much.
I'm very looking forward to our mini series.
Yeah, I'm looking forward to mini series, but I'm really looking forward at the moment to this interview and tonight.
'cause, what we're going to talk about really, I think is fundamentally data science.
But it sounds really cool.
It's got space, it's got linear assets, it's saving companies money, it's it's doing some really neat stuff.
And so I'm excited about where we're going to go.
Do you have any?
Expectations what we should talk about tonight.
Yeah, well, I'm very excited because this is a topic that had so much potential, I think.
And all the use cases that we're going to hear about are maybe just the the starting point for whatever comes with this technology.
What we should do is.
We should bring our guest down.
I'm going to introduce him by first name and let him murder his last name.
To the program.
Thank you very much for having me.
Yeah, my name is a little difficult.
My name is Sven Shiva.
But you can call me then.
That's absolutely fine.
I'm the co-founder and Co CEO of the Satellite data analytics company Life You and I'm super happy to be here today.
So satellite analytics company that that that that sounds like a marketing mouthful.
What exactly is a satellite analytics company?
So maybe maybe just start a little bit with the background.
Over the last few years we've seen an explosion in number of satellites in orbit, and these satellites in orbit produce massive amounts of data about everything on the surface of the Earth, including energy assets such it's overhead.
Lines, railways, pipelines and The thing is that the satellite images which are taken over these assets could be of great use because obviously there are a lot of insights in these in these images.
But what we've experienced is that electrical utilities, energy providers don't really.
Make use of.
Satellite data, and that's where we come in.
We take the satellite data and we analyze it, create actionable insights for which electrical utilities can improve their processes.
And that's what we do.
It's not only marketing, but it.
Sounds pretty cool.
It it did sound cool, 'cause.
You became a guest on the show 'cause.
I read an article about you did null transparency and I thought it was.
So that that is pretty cool.
How do you get into this?
So I've seen you been at.
30 under 30 you're.
How did you find this space?
So really originally I.
Was super fascinated by by space from.
A very early age on and.
Yeah, I always wanted to fly.
To space with my own rockets and whatnot.
But then later on, when I was a teenager and when I was starting to graduate from high school, I was thinking well.
This the space industry back then, was so boring because, like, there were only these huge companies, especially here in Europe. In the US there was already SpaceX, but here in Europe there were only these huge companies.
You couldn't really do anything about it.
It or you.
Couldn't really move anything forward within these companies, so I decided to study something which is a little broader business engineering.
I was always.
Yeah, yeah, pulled back into the space ecosystem because I was just fascinated by it.
And so I started in events years here in Berlin.
For space companies, it's called new space vision.
We also have events now and in the future here in Berlin, as well as our own podcast, new space.
So I met Daniel then in 2017 through this event series, which we started and we just realized that there's not really a lot of people making use of satellite data. And so we decided, hey, why? Why is that? And we said.
Said that we should change that and This is why we started live here.
And then we looked around and thought, well, what's the great advantage of satellite data and it's just it's unlimited scale, you could say they are obviously we satellite today you can run so vast areas of land and what's widely distributed, what is hard for monotone infrastructure networks including electrical grid.
And so we decided, hey, let's monitor these from space.
And this is what we're doing ever since.
So that was your first, your first thought, your first industry that you started to look at to to be use of for them?
Yeah, so to be brutally honest, at the very beginning we looked also at the tourism industry, but this was more to play around with the data and to get our hands dirty.
But when we looked at commercial applications where we we are also seeing a market demand there was certainly the, there was definitely the infrastructure industry and yes that's the first industry we right now we've just raised another financing round.
And are now about to expand into new industries.
But the infrastructure industry is right now the market in which.
And what are those pain points that you're started to to solve for your clients?
What is their problem that you?
Can help with.
So exactly, I mean large scale infrastructure, networks, electrical grids, pipelines, they have common problems.
They have external risk factors.
Yeah, yeah, endangering their effective use and effective operations and and some of them can be monitored.
From space such.
As vegetation ground deformation, third party applications.
Let's really briefly run you through them.
So obviously trees which grow too close to overhead lines can be risk to them they could fall into.
Them and for that either power to just cut.
To cure or.
Wildfires could start both things that we see you.
Want to prevent?
And so right now a lot of like monitoring methods are very manual.
People walk along, so networks fly them or just do it on a cycle basis.
And what we can do is based on satellite imagery, we can identify where trees are.
Our high there are loads.
PCs, they have whether they are sick and we can precisely point to the ones which are too close to the over land and our potential risk for them. There's #1 #2 is change detection exactly if you have a pipeline.
Maybe you want to know where construction sites are being next to them, because obviously you don't want someone build a Hut next to your pipeline or yeah, and so we identified this based on satellite imagery with a high cadence, high frequency.
And the third thing is grant information.
So if you have a pipeline buried buried underground, what you want to do is you want to monitor or.
How that ground is moving and what we can do is from space we can monitor that with millimeter precision how the ground is moving up and downwards.
OK, well that.
Sounds very interesting.
And I'm I'm, I'm wondering you need a lot of data, of course that comes from other sources, I mean, starting with the with the data from the from the satellites.
So when you said early on that that there are many satellites in orbit already, I mean, you don't have your own satellites, so how does?
That work where?
Do you get the data from from above?
Yeah, that's a great question.
And so satellite data is exactly what we buy or get from other providers.
So we don't have our own satellites, we don't have any hardware.
Sometimes microphone is a little sad about this as he's a hardware engineer himself in both.
She added that go to send something into space.
So you can start putting.
Up your wholesale lights some point.
It would do if it.
Wouldn't be that that expensive and who knows, right?
But right now, that's not our plan.
Right now we buy a lot of satellite data from companies which are called planet, which are called Diablos Maxa.
Some of the names you've might seen over the last couple of years, but we also get a lot.
Of data from.
Public data sources such as the European Space Agency's Copernicus.
Program what we do.
Is we buy this data, combine the state in a unique way, and then we need the magic sources in how we analyze the data and how we bring it to the end.
So it sounds like there would be significant amounts of data, right. So I think when I did the pre show with one of your marketing folks I talked about, we recently had a guest on who wrote the book California Burning and it talked about California wildfires in in the problem that PG&E had with inspecting their their power lines, right and as you alluded to.
The way it worked is they either took a helicopter or crew had to go out and had regular intervals that they had to go do this.
But if you're going to do this, be satellite imagery.
The data must be tremendous.
So what magnitude of data are you talking you're working with, and how are you handling the?
So we're talking about terabytes of data for individual clients and we're right now in the process of building up a petabyte archive together with a WS, so.
That also is already part of the answer.
How do we handle their data?
We don't process any data on site, right?
We have a large scale cloud cluster on a WS.
We analyze process a lot of this.
Data and this is really where where how we were able to manage this and the from the one of our company, we were all about optimization, we were all about scale and we also want to serve customers all across the globe as we already do.
And So what?
We have is different data centers in North America and Europe in the Asia Pacific region to really serve the customers closer to them.
And in their home markets, in their home legislation.
So where they have laws, for example, preventing data from from going outside.
But yeah, it's all in.
But you've learned.
How to handle large amounts of data?
How old is your company?
So we've been founded in 2018.
And So what have you learned along the way that's unique to doing these linear assets that that's a unique problem statement that maybe isn't just?
A big data.
Problem, well, I mean so.
So what we do is we don't do any generic data analytics.
What we do is we have concrete solutions developed for.
A very clear target persona and target problem within the energy Energy Company, which is for example vegetation management really where we have built 6 to 7 billion U.S. dollars every year spent on vegetation management and we have in the US alone and we help these utilities so we precisely and effectively.
Manage vegetation next to their overhead lines. And it requires. For example, if your utility of 10,000 miles, 100,000 miles of overhead lines, what we can do is we can.
With so to say, click of a button, obviously it's a little bit more than that.
Get data over the entire network area, analyze it with AI, identify where vegetation is and then generate really actionable task and workers from that, which then can be used by people in their fears who don't even know that the information, hey you have to trim down this tree is really coming.
From satellite data and we've had to learn that that how to how to translate satellite data into actionable insights on large scale very early on as our first customer has been the general operator who's biggest transportation company.
And our first first the idea was to analyze the entire 20,000 miles or 33,000 kilometres of the railway network regarding vegetation. And so it was a yeah, drinking from the fire horse and directly from the beginning.
I I can imagine and that brings me to my next question because I saw that part of your solution is also an app that your clients can can use when looking at their own assets.
But so for your analytics, for your AI, you only you not only use the data from satellites, but you also use data from the client, right?
Because you need to know where do this, where do the railways go or where is the grid?
So I I was wondering how do you handle this?
Probably very private or sometimes private data from your client, because not everybody wants to share.
This is where my pipeline is buried in the ground, or this is where.
My, my railway.
Goes well, maybe for red ones.
Yeah, 100% so.
This is this is obviously like data privacy.
Data security is.
Super important to us.
And So what we will put a lot of effort and an emphasis on is data security and data privacy within our system.
So we've run through certification processes to assure data privacy and data security within our systems and we make we take put a lot of effort into making sure that the data the customers share with us.
Stays within the system and is only shared with the customer and no one else.
And so we receive information about where their customers infrastructure network is.
So obviously you've touched on pipelines, they maybe the infrastructure network, the least amount of people know about where it's running?
But generally, for vegetation and overhead lines in railways, oftentimes you can see where the infrastructure network is, but regardless of that, knowing that for the entire network is critical.
And so the customers share this data with us and we make sure that it's not getting outside.
One of the interesting things I think that you and your colleagues had mentioned is in the vegetation you do things, maybe know the tree height, the predicted growth and you help prioritize or create maybe trouble tickets and help enterprise manage that.
Is that true?
Is that to understand that correctly if some of the things you could do?
That's absolutely correct, yes.
And I guess, so how does that work in today's world?
What is the value to the business?
So I bring in a company such as yours, you take some satellite imagery, you you monitor my entire linear asset and you give me the vegetation.
We know vegetations a problem for power companies.
You said fire outages and all the other things.
How does that work? Do you, do you use machine learning's to some predictive analytics what what, what I expect is your customer and how is that?
Gonna help my bottom line.
And so you imagine you're an electrical utility. And So what you have is 10,000 miles of overhead lines. And right now you have a vegetation management budget spent on this where you have an annual.
Budget which is typically spent on a cycle, meaning that.
In year one, you look at one part of the segment of your network, in year two, another part in your free third part and so on.
And then obviously you would have some certain budget set aside for for like emergency cases where you have to trim down trees which have already fallen into your network or with someone called an emergency case so.
This is the status quo. You would have people driving there, you would have subcontractors streaming down the trees. But but really the main thing is that right now you work on an assumption that the assumption that you go from A-Z in that cycle and that this is currently, that's this the best method.
Which you have at hands to, yeah, make sure that no trees grow into your overhead lines.
Now The thing is that we we see that the world is becoming more and more dynamic.
We see climate change through that.
We have changing weather conditions free, changing patterns of vegetation growth and storms obviously becoming more dangerous.
Because they have species coming out in from other parts of the globe with completely different growth patterns and so, so this cycle is really under pressure in terms of it's it's usability in.
Real life still, what's the alternative?
If you don't have information about what the status of your asset is, it's super hard to make out where you should act now.
And this is really the gap which we're filling with life here.
We are identifying, we're helping the utility to identify precisely where actions have to be taken, not just like because it is now 4 years until.
Have been to this place last time, but because we see on a satellite image that there's an action needed there and with that we have to improve the efficiency of these processes.
We we have to be more effective in such a way that we have to prevent power outages and we have to prevent any kind of danger to the network.
And then obviously all of that in the end ends up in the bottom line, right?
How much money do you have to spend?
If you are more efficient, you maybe have to spend less.
Maybe you want to spend the same amount but they want to achieve.
More and if you're more effective and can prevent any damages to your network, you're also safe because you don't have any CapEx spending, right?
And maybe you want to.
But this is what we're doing and this is how we're helping the bottom line as well as the end customer in the end, which has more reliable network.
So you started this company four years ago, if I understood correctly.
So it's pretty new and I mean everything you you do hasn't been here and hasn't been possible like 10 years, 15 years ago.
So I was wondering, how do your clients react?
I mean, of course they have to start trusting what you do, so maybe I could imagine.
They're not completely going away from their cycle based maintenance, but maybe just try.
Your solution for a part of their their assets or or maybe just them starting to to shift from from the cycle and these is that what tablet happens?
So first of all, I think like podcasts like this are so important for the energy industry just because the energy industry right now is going through a transition which is which is without parallels, right?
We have climate change, which we want to fight, we have.
Of changing population patterns across the globe like and then we have exactly other or different kind of energy transition.
So it's it's it's incredibly important.
That's the energy industry and I've seen this every day.
UM is changing, becoming more and more innovative and obviously taking a jump from cycle based approach completely to a satellite based approach, which sounds like it's coming straight from science fiction.
It's a it's a big leap of faith for a lot of employees at utilities.
So we try to help them with making a decision and making also decision, which they can argue internally.
And So what we.
Do is we?
Run in our sales process.
And we run a PC.
At the very beginning where we take note maybe the entire 10.
1000 miles of their overhead.
Fine, but we may take 100.
Wasn't my thing and we we analyzed that.
We we provide the insights to their customer base to to to their employees.
We have a validation process, we we on scientifical measures really show on site how good our analysis is and this gives them the proof and the confidence to roll that out and that's what we're doing.
And more technically speaking, how do you train your models because there won't be a data set.
This is all the fallen trees on networks in the last five years here with, I don't know, latitude data.
So how do you, how do you train those, those models with what?
Yeah, we obviously this is this is part of our secrets or.
So without giving.
Without giving too much information away, training our machine learning algorithms and creating the necessary training data for that is is one part of the where we put a lot of effort in and we have innovative, I think a lot in the last couple.
But yeah generating training data is not easy and not easy to come by but but we solved that and we believe that we have sorted also on large scale and there's a human element to it for sure as we don't get around first creating the the material with through training machine learning.
But we've optimized as much as we can, yeah.
I buy that you you're automating that.
You're learning and getting there.
We spent a lot of the interview so far talking about vegetation, but that was only one of three components that we started the interview talking about.
So maybe you can shed a little light on maybe what you do with pipeline and what the value add is?
I mean, obviously building, uh, huts are pretty easy comparing or easier I think, than than some of the other uses.
What are some of the more advanced capabilities you have for, let's say, a?
So exactly I think like so, so.
So we have two categories, two additional.
Stories of our.
Offering wireless change detection.
The other one is grant information analysis I.
Will dive a.
Little deeper into train detection first.
So within change detection we.
Kind of two two segments, right?
One is the rapid change detection analytics, which is really important when you need direct insights about your infrastructure grid within hours after an event.
And we call this new real time monitoring and we provide these two electrical utilities and other utilities such as railway operators or pipeline.
Operators where they will, we help them to precisely identify where something has changed alongside the infrastructure network.
OK, right after a storm.
For example, we help the generator operator to identify precisely after a storm where trees fallen onto the network within hours.
Right now this is not possible because right after the storm it's it's difficult to fly.
Maybe you don't have to write planes and whatnot, so we can do this from space.
And then there's another aspect to change detection, which is.
Slower changes such as construction sites, so.
Obviously if you.
Have a pipeline buried underground.
Some some people just don't know.
Other people don't care if there's a pipeline and what they do is.
They build something next to it and right now pipeline operators have the obviously interest and also the obligation to know where that's happening.
And so right now they fly on a cycle, their pipelines to identify where things are being built next on top of these.
Pipelines to then stop these activities and.
What we do is.
Based on satellite imagery, especially for more rural areas or as an addition to the plane flights, we provide insights on a frequent basis identifying where changes have occurred next to the pipeline.
That's the change detection kind of piece.
Do you also take drone data or plane data?
I mean it's satellite is just data at the end of the day, right?
So do you take imagery currently from drones and planes as well and use it or only satellite?
At the moment.
So, so we have used and analyzed big amounts of satellite and of drone and aerial data.
The thing is that we are all about scale and we are all about bringing really a, yeah, a good solution to the end user.
The thing with drones and planes is that oftentimes we work with local vendors or with local.
Kind of setups.
Every plane with a camera has maybe a different camera attached to it.
Every plane with a camera is maybe flying at a different altitude.
Well, everyone who's like trained in machine learning algorithm knows that obviously all these strangers make it more difficult to be able to produce reliable results.
And so we say first one what we do is we focus on.
Satellite data analytics, but yeah, we have any and vast area and drones and we will continue to do so in the future, even more so than.
We do today.
OK, from the third branch the the ground deformation is more about.
Yeah, natural catastrophes or what what could I could I understand from that.
Yeah, it's more about natural changes, absolutely.
So what we're looking here at is monitoring how the earth underneath of railway tracks and in pipelines and other types of infrastructure such as electrical lines, is moving up and downwards.
This mainly has natural reasons, for example.
Wash out from from waterways or.
Geological reasons, but also could be the mine, an underground mine underneath, uh, an area could be a reason as you see it in some historic and mining areas in in all over Europe and North America where you have underneath mines which people forget about it and suddenly the earth is moving down.
What we can do and this is I think it's.
Super incredible satellite.
Imagery radar imagery allows you to monitor.
The movement of the earth up and downwards with millimeter precision are based on something which is called.
Insight analysis and this is what we do and we help the operators of pipelines for example, to identify where that's where that's happening and to make sure that it's it's not really affecting the pipeline or the railway negatively.
OK, monitoring trees and vegetation and monitoring the surface of the earth going up and down is probably a big, uh, huge different approach that you have to take and maybe you can shed some light on, on the the different approaches for the different branches that you're working on.
You're absolutely right.
When saying that they are different in terms of the the analytics which we use, right?
So we use VC.
Machine learning as well as other types of analytical processes for both these these things where it's all the same as OK, well, how do we, how do you make use of the data, how do we access data, how do we scale processing and how do we bring this to the end user?
We have a unified front end web app, mobile application, Arcgis integration, SAP.
Integration and so forth.
Which users can really make use of the data?
And that's one of the big gaps surfing in a lot of AI and analytics software.
How do end users really make use of the data?
So the middle part where it's about processing this is different and there we.
Have build up unique.
And proprietary software applications to monitor machine learning vegetation with machine learning, change detection and machine learning and other types of analytics, ground information analytics.
And we have separate teams for this.
Does that dinner question how big is your team with everything that's going on?
So we are a little bit more than 100 people across 33 offices. We have one office in Latvia, one office in Berlin and one office in North America.
OK, I have another question.
Uhm, because with the satellite imagery and all the potential that.
As I would think that your clients are able to to monitor some regions at a lower cost than they used to and maybe the cycles that they chose to to use before in those remote or difficult or yeah difficult accessible areas were lower than they should or not that they should be, but.
Now they have the possibility to to monitor this much more closely and more frequently.
And my question is, is there any change in behavior of your clients that you have experienced because with the lower risk or the better risk management that they can do with your solution?
Maybe they're doing something different when it comes to investing in those areas or something.
I mean, this is the big promise, right?
And this is what we're what we're working on together with all kinds, right?
To walk this, this, this path?
As we as we said we're four years old and we are in roll out when we have customers using our solution.
I mean, then taking exactly this, this next step of where to make investment is the one which comes afterwards.
And this is what we're going to see in the next couple of years.
But as you know the energy industry is sometimes moving a little slowly.
So we're we're we're still working on that together with our clients.
I I see the transition taking place.
I I see new and emerging technologies much like yourselves that there's a couple of competitors that do similar.
Things to what you do.
I guess telling me that you're four years old, one of the things that I would think a data set would give you is a nice trend analysis and some tools that we don't necessarily have today and maybe predictive tools.
So if you look over multi years of data, how does that change the processor?
What are you anticipating or hypothesizing?
That this data could do so, let's say at 10 years of data from you guys.
Is there things that I'm going to know that I don't know today from the old process?
Or maybe with the scale of data that a human can't just pull out of the data because it's just too much data?
No, I mean.
I mean even from the one you, you will be able to drive change right so exactly from day one as to go to come back to the example I've just given where you have four year cycle right and you use live view and you get every year you could inform.
You get it four times compared to your two initial kind of process, right?
And this drives behavior.
This drives change because it really helps you to make a decision where to go today and what to do based on on on this grid wide overview, if you have 10 years or even like 2 years, three years you can really see.
Maybe patterns showing up in terms of where vegetation has a bigger problem, or where changes are coming more often and we want to maybe also move to a different kind of solution.
Meaning having drones or having person on.
Site for the exact.
Area taking it, making sure that no one is building a construction site or spilling a heart.
In that part of your pipeline.
But but these patterns become visible from day one.
Then driving conclusions from that, or we see sometimes comes with time.
But we try to give our customers all the tools they need to to drive this conclusion, which sometimes require context, regional context, context.
Sometimes just people on site can have.
So we have.
A set of web app and mobile app and other integrations which really gives the insights of satellite data into the hands of.
Uhm yeah, normal people which are working on site which are really like making the the maintenance or conducting the maintenance alongside these assets which have nothing to do with satellite data, which have nothing to do with analytics and we we we make, we enable them to too.
Yeah, be more efficient for these insights, which by the way in our web app, in our.
Mobile app you.
You wouldn't know that there would come from satellite data, and it could come from a different data source.
No, but I get that, right?
I mean, that to me makes all the sense in the world.
You put the tools in their hand and they can work differently.
But I think of it this way.
As a technologist and spent my career in technology is usually a new technology becomes fundamental.
Fundamental to, let's say, the next killer app, right?
So when you start this company, you say, hey, great, we're going to solve this problem and have lots of data.
Now we've got this huge corpus of data to work with.
And you know we can start doing some other things that maybe we didn't anticipate.
So four years in you probably have you said you're going to petabytes of data or you're going to have a lot of things to work with.
Is there a killer app in the horizon in your vision from being four years in?
Or is there a killer use of the technology that it wasn't designed for that you unexpectedly or come across so far?
I think nothing unexpected.
I think every day we are, we are we are facing another problem or another area maybe in the energy industry, but also outside of that where we believe that satellite data could be a great part of the solution to solving that.
Speaking about like for example the forestry sector in which we currently are looking into the insurance sector.
The logistics sector and all of that where we where we see solutions that we see how our technology which originally was built for the infrastructure sector could be very helpful.
And so this is also like the goal of our company long term to monitor a billion assets across industries.
How do you organize the flow of information from your clients coming back?
Because you mentioned you have the mobile app, so the people on the ground really checking on, on the incidents that have been raised and they have the app and they put in what they see, how is that organized?
Because I can.
Yeah, maybe that's just it.
Not going to be a free text field, but yeah, how to to organize.
The data to make it scalable and make the flow back to your to your validation and your models.
Here we have a couple of UI designers and UI developers which make sure that it's nice and handy.
Another free field.
So we have questionnaire, we we have the ability to put in pictures, voice messages, videos and everything which which makes sense and we store it connected to obviously DS app which they were looking at.
I I love to use spatial data because everything is connected to.
Everything else through a location, right?
And we also like obviously have this feedback cycle where we have satellite data top down and bottoms up with the field view and we connect that.
On a map.
Through our web app for mobile app which.
Provides this on site data and.
This is how we do it.
And on what map data gives based data do you you based?
I mean usually everybody knows Google Maps, but you know you you need to probably St data to to be able to to to see the entire picture.
So what what is your source there just for the?
Yeah, like the normal mapping.
So, so we just use something similar to Google Maps, which is Mapbox, which is a mapping framework used by, for example, Uber, which is just a baseline map.
On top of that, it's satellite data analytics which we create ourselves.
So far it's been interesting.
I guess the on the business.
Side let's talk.
A little bit about the business.
So you solve a business problem, you're you just said you went through a round of funding.
Where does this go from here?
So are you in that growth mode?
Are you still in the vetting mode?
Where's the company at?
So, yeah, exactly.
I think that we're absolutely in the growth mode.
We have believe, we believe that we are and we see that every day.
We scaling our processes internally.
Because our customers are asking for that, because we see the customer demand, we see the customers projects coming in.
We have customers on every continent of the Earth.
And this just shows the great advantage of satellite data that it's really largely scalable and this is this is where we are at right now.
We've, yes, we've just closed the financing round giving us the ability to accelerate their growth path and growth trajectory and we are we're now expanding our teams.
So if anyone out there listening to your podcast.
Is looking for an exciting job in sales, business development or any tech position with a lot of positions on.
Well, great for the commercial, I guess.
But what I wonder then is what's the ecosystem?
So we talked about your app and you said, hey, we're going to connect.
So if I've got a trouble ticket system or or some other things, what kind of partnership and ecosystem do you have to have to really get full advantage of your?
So the good thing is that we implement with like the.
The factor standards of a lot of the energy industry is at least that we as we perceive it so.
With an S3 integration actually as overseas, one of the tools being used by most energy utilities as well as other companies, we have an integration there with an integration into SAP, which is also quite, quite widely used in the energy industry and what we have is a dedicated team providing a unique APIs, right.
If you are.
Talking to an energy operator or energy utility.
Sometimes they use out-of-the-box solution as mentioned. Sometimes they have their own home build stuff. Then you still want to integrate.
But we we have a team which which can sort these things out.
So we need.
To but even without any internal systems, you can take full advantage of our solution through our own applications.
So get the integration now.
Now, the kind of dumb question that comes to mind is earlier you mentioned that typically, let's say, there was a weather emergency or some sort of catastrophic event.
And you said, well, we can get answers in hours.
I believe you're in Germany, which is not the sunniest country I've ever been in.
Uh, not always clear sky in hours.
So how reliable are satellites in in in that situation?
So I get it on a perfect day that there's a problem that happens and it's a clear blue sky day, no problem.
What's the percentage of coverage and how quickly can you get there 'cause that to me could be a conceivable.
For real time data if it's near real time.
Less of a problem, but if I want to.
Monitor after a snow storm, there might still be clouds in the.
Sky or something like that.
100% and we just monitored through the snowstorm this winter. So the things what we use.
For that is radar.
Radar imagery doesn't have the disadvantage that it.
It's being blocked by.
Clouds, but it can see through clouds.
And this is a beautiful thing about that data source.
It comes at a price, yeah.
But when time is of essence, so radar data comes to your help, you could say and this is what we do, we procure radar data which is yeah, a radio wave which penetrates clouds which in the anti satellite image coming from radar satellite looks very, very different from an optical satellite.
But we take this set to end.
Process it and in the end translated into actionable insights, so we don't have that problem.
For the other types of solution which I've just mentioned, we mainly use optical data.
As it is provides more context and more able more ways to analyze than the radar data.
But as said under contrary radar data is capable of providing new insights no matter what time of day, no matter what weather conditions.
So for the emergency case.
In an emergency case, yeah.
So it seems like pretty cool stuff.
You haven't let me down based on my expectations so far of what we're going to talk.
About you know, we've talked about satellites, imagery, projections, all all the cool things you can do with it.
Guess what I'd want?
To know is what are some of the lessons you've learned or what what what did you think what assumptions did you have going into this business that didn't prove out like what has changed from in starting a business we all we had some assumptions.
Four years ago.
Right. Yeah, sure.
What were some of the erroneous assumptions?
It can't be that.
Hard to sell into the energy industry?
Well, that was the first assumption.
I felt I was proven wrong.
It's very hard to sell into the energy industry as is.
Maybe some of your listeners know maybe sitting on either side of the table and it's kind of discussions.
So selling into the energy industry is something which is painful, which takes long time, but obviously it's also very rewarding once you have worked with a customer or once you are working with a customer.
Because customers in the energy industry also are very open.
They they love to cooperate on also new developments, but yeah, selling in the energy industry.
And the other thing is, well, satellite data isn't can't be that hard.
While it is very hard if you're talking about analyzing data at scale and in life, here we were at day one about automating and scaling satellite data because our goal is to support all infrastructure grid operators globally which have to maintain.
In the end the.
The the modern backbone of our modern industrial society.
We don't support them with satellite data analytics.
And I said long term we want to monitor a.
Billion assets globally it.
Only works if.
If you put all your emphasis on optimizing and scaling and that's quite difficult and we at the beginning we are a little bit, yeah, not even thinking well how we just need a couple of engineers then we can do it.
I said we are 100 people, most of them being.
Engineers and they have.
Cracked the code when it comes to satellite data analytics.
But it has been, it has been a journey to to get to that point.
I totally believe that.
And looking at the other side of the table, your clients, what is their feedback?
What is that?
Was their biggest surprise?
Maybe even started working with your solution?
Yeah, I think at the end, you already pointed to that at the beginning.
Electrical utilities or infrastructure grid operators are rightfully cautious about like what kind of solutions they want to work with in the end.
I mean, they they make sure they hear the lights on and we can record this podcast.
But I think like once they've like came, came over this first barrier, his first hurdle of trusting the solution enough to to test it out, I think they very, very quickly were very, yeah, amazed and surprised about like the large scale kind of insights they've gotten.
So we've always gotten very positive feedback there there.
Head of manager, head of operations at Dodge Ban, who who has a budget of multiple billion dollars, said.
Well, the solution set light of life here helps us to to make sure that we can.
Provide better service to our customers that our operations are better.
Our electrical grid customers are very open in terms of communicating how, how well that works out.
So I think like we have a couple of.
So our customers are generally very, yeah, excited about the opportunities which they already see.
Life in action and we'll see in the future.
I don't have.
Any other questions?
Selena, do you want to ask a final question as we wind up the show?
No, I think we're we're all good.
That was very interesting.
For me, I I think the the use of data is really powerful.
I agree with you working in the energy industry that it can be hard to introduce in it can be hard to get people to change.
From legacy process.
But but what I do see is based on the current environment and where people are a need to to do this.
And, you know, vegetation, I just know from from an energy perspective is a huge problem and it's a costly one.
So if there's significant savings and better predictability and you can.
Automate and reduce the cycles.
I can't see that not inherently being successful at some point once it's vetted out and as you've got, you know, experience and get customers vetted out, I would think that that would take off and be important to the industry.
I I want to thank you for sharing your story.
Definitely enjoyed hearing about your journey of how you start a company and getting.
The space, you know you, you say.
Not everybody can get into space.
I think they've even had a neighbor that has a space company that does low Earth orbit satellites.
So I I think it's probably more popular than than maybe you think 'cause it's it's it's big.
Business as well.
But it's really cool to see using the data.
But to me it's it's more a data story than just a space story.
'cause it's just images.
That you're working with and doing data science.
Here's what I think your company is from the cheap seats.
Thanks very much for having me.
Was a pleasure talking to.
And exactly, yeah, I said, I think it's great.
What are you doing here with the podcast?
And I believe that the energy industry is going through such an interesting change that it needs a lot of smart brains to work on the problems we.
Have ahead of us.
Well, thank you for being a guest.
What we will probably do is in about 6.
To 12 months.
We'll bring you back and see how this is all playing out.
If you're willing to.
Come back on the show because.
I'd like to see.
We do a segment called.
Where are they now?
We bring guests back about a year later and see see what's transpired in their predictions.
So last question I have for you is 5 years from now, where are you guys?
So five years from now, that's a great question.
So what we where we are in five years we had a.
Folks, the the leading of observation and analytics company on the Earth.
We have major offices in North America and Europe as well as in Asia Pacific.
We will have expanded into more than just the energy industry, but having expanded to a handful, maybe a dozen of different industry and what.
We have done is we have opened up the architecture which we have to other people building on top of that one to bring satellite data analytics to customers across all industries.
I I wish you well on the journey.
I look forward to touching base with you.
Thanks again. Thank you.
For our audience, this has been another episode of Insiders Guide to Energy.
This kicks off our data science focus.
Tune into the data science mini series.
You're going to hear a lot more content on data science and all the cool things you can do with machine learning, AI and how it can help the business and where it's going.
We'll talk to you soon and talk.
To you again next week.