Adam Sroka (00:47) Hello and welcome back to the Hypercube podcast. I’m delighted today to be joined by Markus Dirnbacher Markus, would you introduce yourself please for the audience?

Markus (00:58) Thank you, Adam, for having me. My name is Markus Dirnbacher. I am working for Innova Renewables as a Senior Operations Manager and very excited to be talking to you today.

Adam Sroka (01:11) So you’ve got quite the impressive career in energy looking at quite a few roles. Could you just expand on why did you pick energy and talk through some of that career journey for us please?

Markus (01:23) Definitely. I think story time. It goes back to… I’m from Austria. And as you imagine back in the day, we had lots of snow, we had frozen lakes everywhere. And I really enjoyed the outdoors. But over the time, I realised that we’d gone down from a meter or two of snow in the winter to just a little bit. The lakes weren’t frozen anymore. So we

couldn’t go for pond hockey with my mates and these sort of things. So I started to get interested into what is actually happening. And when I got closer to university and started to think what I want to do with my life in the long term and what I want to spend my day to day with, I was drawn towards energy and thinking about how I can have the biggest positive impact with what I do day to day.

Led me towards the renewable energy industry. And I started in my Master’s work experience at a solar O&M company where I then ended up to stay for seven years. And that was the kickstart basically.

Adam Sroka (02:40) Amazing, very cool. It’s funny when you see so many people in energy, I say this all the time, but because I’ve had the benefit of working across a few industries before I settled on energy myself, especially as a consultant as well, you see a lot of other businesses and unlike anywhere else I’ve ever seen, you get a wealth of really talented people that could probably go off and get like triple, four times their salary working for like banks or big tech and things like that.

But they stay for the kind of reasons that you’ve cited there like they want to actually do something with their career and with their life that has like a real impact not just optimising shopping baskets for some like massive multinational shop or something and I just think it’s really yeah it’s always really interesting to hear story time so yeah thanks very much that’s really cool I’ve never played pond hockey so if I ever get to Austria you’ll have to teach me.

Markus (03:35) Yes, yes, if we have ice here.

Adam Sroka (03:38) Yeah, if you have ice, I’ll bring my own.

What about Innova then? Could you just expand on what Innova do and who they are for some of the audience that might not know?

Markus (03:48) Innova Renewables is a vertically integrated renewable energy company. And we have an internal highly skilled team of engineers

site delivery professionals and then operations where I come into the game. And we see these sites through from planning to operations. We have certain different partnerships in the market that help with delivering these projects. And yeah, that is the general business case of what we are trying to do at Innova.

Adam Sroka (05:15) Where are these assets?

Markus (05:17) Most of them are in England. There’s a couple of them in Wales, a couple of them in Scotland, but everything is UK based.

Adam Sroka (05:28) You must have like a really well-oiled machine that’s kind of gone through the motions of developing so many sites and assets and handing them over to operations. Is that the case? Like have you really perfected that process?

Markus (05:47) It is a lot of assets, saying that, we are really starting up now. So we have roughly 100 MWs in operation at the moment. This starts of course with our development department. Yes, they have a very good process. It’s a very well-oiled machine to have all these critical

permissions in place to get to ready to build. So they do the very good job upfront. And then we have our engineering department internally that work with the EPCs and our delivery project managers that then go alongside the EPCs in constructing these assets. Everything goes well, it comes to me at provisional acceptance. And then we are in for the 40 years of the…

Solar Lifecycle.

Adam Sroka (06:46) And yeah, but the reason I’ve kind of wanted to talk about that was is there there’s must be like this massive handover piece where, and I’m thinking like data dump where you’re getting like checklists of things gone over and handover of controls and systems and reports and things but like yeah that must be quite an interesting time when that’s going on

Markus (07:10) That’s the advantage of having an in-house development department and all the project management as well is that this structure of data is set in the beginning and it grows over time. So from my past, it often happened when you were handed over a site that half of the documents were missing. The big advantage here with us is that we basically structure in the beginning of every project.

Adam Sroka (07:21) Hmm.

Markus (07:40) And it depends a little bit on your route to market. For example, if it is a CfD site, you have to deliver certain documents at, I think it’s when it comes to PAC, to prove that you have built the site that you were promising. So we can set the targets in the beginning of what data will be required further down the line. And basically then tick off these check boxes if once we get to these.

And have it all in our folder structure ideally. And then it’s just a matter of from the project development manager giving it to the project manager, then the construction project manager, and then on to me. And we all work on the same folder structure to have it all nice and safe and readily available.

Adam Sroka (08:29) And like for, we’ll have a few, quite a few people in the audience that probably have never developed a site or looked at that well. Like I’ve never developed a site, but I’ve consulted into these things. And that’s not like, it’s not like a few files, is it? We’re talking like thousands of contractors and contracts and stuff, right?

Markus (08:43) Yes,

Thousands, thousands, it’s gigabytes and gigabytes and gigabytes all over the place. So you have to be structured and organised. That is critical for even the success of the project in the end. Definitely. It’s not an easy task.

Adam Sroka (09:01) No, and that, guess, yeah, that structure, the way you were doing it, that becomes IP, like, in and of itself. Because it baffles me, so we’re actually in the early design phases of developing, a piece of software that’s going to look at this, because there’s nothing out there that’s bespoke for this use case, like developing sites for the energy system. And we talked to so many people that maybe aren’t as mature in their processes as you are.

And it’s just like a hodgepodge of email and PDF and random ERP systems.

Markus (09:35) Exactly, and it’s a big risk. If at certain stages of the project, stakeholders leave and the data is in their inboxes, it can get lost. Yeah, and that is a problem. So you have to think about that. How you handle this data is very important.

Adam Sroka (09:45) Exactly.

Okay, so right, let’s move past that a little bit. So Senior Operations Manager, could you talk us through like a typical day? Like what does that mean?

Markus (10:04) There is no such thing as a typical day. Every day is different. The thing with operations and asset management, what I’m doing at Innova, I’m involved in many different departments. So that starts with the planning team when we talk about, for example, landscaping, because usually you start once the site is constructed, then you start with all the…

Adam Sroka (10:06) Crack it.

Markus (10:30) tree planting, hedge rows and these sort of things. And it depends on what your planning department promised to get the site through planning. And if that is also ideal to do or practical to do, which sometimes it isn’t and we need to slightly change the plan. So that means I have input already in very, very early stages of the project. It goes through to…

design decisions of how can we optimise a solar farm thinking about the 40 years of its life cycle. So there’s sometimes questions in this area. It goes to contractual management of negotiating different sort of contracts with suppliers or the O&M contractor and that sort of aspect. And then it starts to go into the performance of the solar assets. If there’s issues, if there’s

performance issues, contractual issues, other sorts of risks. And to summarise a day, think the overarching constant is it’s problem solving. There is problems constantly, but nothing that can’t be solved at the end of the day. All for the good of the asset and for generating renewable energy.

Bottom line, think that the best description of day-to-day is problem solving.

Adam Sroka (12:00) Yeah, very cool. It makes me wonder, do you have a favourite site and a least favourite? Is there like, are you allowed to say or? Yeah.

Markus (12:07) Definitely.

I don’t think I’m allowed to say, definitely, think if you ask any asset manager or even in construction, there always is the one site that is not going absolutely to plan. But it’s again, problem solving and getting the site to also live up to expectations. And saying that

If you get these kinds of problems, it’s very interesting because that’s when the learning happens again. If you’re faced with stuff that you haven’t seen before and you have to go into the deep dive and really figure out what’s happening. So that is always a very rewarding process in the end when you get there. Might not feel that way whilst you’re stuck in the middle. But once you get there and you have the desired result, it’s always very rewarding.

Adam Sroka (13:04) So we touched on management of like gigabytes of documents. And in my experience of speaking to people on all sorts of roles, there’s a lot of like looking at either project plans or going over specifications or reviewing contracts from like a long time ago where something was agreed. Just interested to know, are you starting to see anyone, is anyone like overlooking at using AI for some of that

text crunching or some of that workflow, what kind of use cases are you seeing that be beneficial?

Markus (13:39) Yes. So what we are doing now is overall in the company, we’re building a data lake. In that data lake, there will be a data warehouse for operational data specifically, but the entire company is looking into that. So if we think about what AI is doing and how we can leverage it, I think it is best explained with it eliminates process.

You have a problem and you get a solution. What we need to come around with is, do you trust your solution that much that you can let the AI go and do its thing or you don’t? And that heavily depends on your dataset in the beginning. The quality of data, validity, continuity, if you don’t have data gaps and these sort of things in every part of the business. And then…

I think it is very important that you sit down and have a good think about what your AI should actually solve. I don’t think that the overall process of, okay, let’s do AI. There’s a dataset. Let’s see what comes out the other end. That won’t work. So you’ll have to have a very good idea of what you actually want it to do. And then you can start to look into automations and measure your success with AI.

If you have certain

processes, mainly admin. And that’s for me the goal of AI in the short term, to get the admin off the table and let AI do this kind of work. But yeah, we are quite active in this space starting this year really, to look into the details and define processes that we could see AI working very well.

Adam Sroka (15:32) That’s really interesting and it sounds like you’re quite a bit further on than lots of the other people we talk to in this kind of space because although there was this big hype around data lakes and that kind of Lakehouse storage and stuff like that years back, it doesn’t seem to have made its way to the energy sector, any of that discussion. And lots of people, when we talk to them, it’s all quite, it’s like a new concept. So it’s interesting to hear that that’s,

that’s under development. Absolutely love to get under the hood and have a look at what you’re doing and how that’s being done. But maybe now’s not the call.

Markus (16:02) Yes.

The details, yeah, we would need to talk to other people as well to get under the hood. But I think what you say is quite right. The energy industry being late probably is because we had these 5 MW Feed-In Tariff sites in the past. They just made the business case. The bottom line was good because the price for the electricity generated was really good.

There was no absolute need to go into the details and kind of figure out another 0.5% of performance somehow. But now that the energy landscape is changing, it becomes more more important to optimise performance. And that only is possible if you have that data and you can look into where to adjust and get a better outcome.

So I think that might be a reason. It’s just maturing in the industry, I think, as

Adam Sroka (17:05) So that is really interesting, I think.

You’ve because you’ve talked about like using the AI piece to do the admin. And I think that’s in, the kind of the development operations, the early stage of a project lifecycle actually like that is one of the most valuable, easy wins like that we could get after and where the current tooling and because this technology is like accelerating and actually the industry doesn’t really know what it’s best at yet. Everyone’s trying to figure this out.

So before we hook it up to like controlling these big super expensive assets and letting it take over like actually the admin piece is wonderful. But I’d be interested to know is that that vision of what you want to do with it? Is that the Innova strategy for it or is that just like a personal because that’s your domain or has Innova got sights on like other big wins?

Markus (17:59) So the Innova approach to AI is we are happy to look into applications that make sense. I’ve taken that approach into my area of expertise, which is, it starts even before getting huge insights into how your solar farms perform. So first applications would be the data gathering. We have different sort of assets. We will have different inverters,

and different data streams coming towards our data warehouse. So the first task is to normalise all of this data and have it readily available to analyse. And then the first applications of AI and machine learning start. So what happens if you are missing data from pyranometers? Usually you have more pyranometers at the solar farm. And what happens if one of them doesn’t send data? So historically,

you might have taken satellite data and manually inputted that. The benefit of that is somewhat questionable because of the accuracy. But now with machine learning, you could look into calculating the irradiation data for that one sensor, for example, by looking into the history, how it reacted to the other pyranometers on site. And the AI input could be to automate that. So whenever

Adam Sroka (19:24) Mm-hmm.

Markus (19:25) this kind of fault happens. Your dataset is not interrupted. You don’t have a gap. You won’t have an error coming out at the equation at the end, but it’s backfilled with actually calculated data. And then the next step would be what happens if the pyranometer comes back? It was just a communication issue and the data is available. Do you replace it? Do you keep the calculated data? And that would be some practical applications in the first

data gathering aspect before we even go to, now let’s combine inverters on advanced analytics and how they perform against each other and get into predictive maintenance in a sense of, what’s gonna fail next? So that’s what I meant with you have to know very well what you actually want to solve with AI and do tiny little steps.

And get your automations in before you take over the entire big, big picture of, I want to do it with everything.

Adam Sroka (20:32) I couldn’t agree more. I totally think that that is the way, I gave a talk yesterday about this very thing, how there’s a lot of FOMO-driven development going on and people are saying, all right, we need an AI and so they invest something. And there’s an illusion of progress, like really fast pace on the technology front because these tools do that. But that runs this massive risk of

doing the wrong thing too quickly or whereas those small use cases prove it and iterate with like fine, nice self-contained, easily verifiable, easily testable modular things that will scale. Then that slow steady march towards like a higher value. I think it’s just a really good way to tackle this. And I think actually just the way you talk about that as well, like pyranometers, communications issues, transformers, I think it doubles down on the

Markus (21:02) Heh

Yes.

Adam Sroka (21:30) The hunch I always had about the energy sector is that you can’t just go and farm this stuff out to random tech companies either, because they just don’t, it’s so complicated. Like what you’re talking about is so in depth, so complicated that it would take someone that didn’t know the energy sector quite a long time to get up to speed with even understanding the question you’ve asked of them and stuff like that.

Markus (21:56) Yes, I think it is. You need highly specialised people looking into that because you need a little bit of background of what actually happens on site and what causes these issues and then to have a bit of knowledge. Is it just a communications issue or is the equipment down to draw your correct conclusions afterwards? So yes, definitely you need a background in knowing how solar actually works or BESS

will be the same. To draw the right conclusions then, yeah.

Adam Sroka (24:01) And you six of your solar projects were recently rewarded CfDs in the AR6 auction, right? How does that kind of support your ambitions and what does that mean for those projects?

Markus (24:13) CfD is just a brilliant process, I think, of giving stability to a developer. So with solar, the value of your project heavily depends on the power price. So there are certain times of the year where, or the year in a bigger perspective as well, where it makes sense to build a project because, for example, the module prices are low or transformer prices are low and power prices are high.

Sometimes the power prices are low and the equipment prices are high and it doesn’t make that much sense. What the CfD does, it gives you that stability of a strike price for that project for lifetime. And the government guarantees you if the power price is below that strike price, they’ll pay the difference. If the power price is above, they’ll get the upside. It just makes projects bankable.

And it sometimes makes or breaks projects. It could have an effect of a project that might have been dead before to get it back to life because suddenly you have that bankability and it will deliver these revenues over time and then it makes sense to build it.

Adam Sroka (25:11) Mm-hmm.

Yeah, and it unlocks like a different…

group of investors as well, because some like the big pension funds and stuff just won’t go near the volatility. They want that long-term security and it can really, it’s like a non-linear process. It’s like that’s even a slight change to a little bit more stability, unlocks like a load more funding and makes the whole project much more viable.

Markus (25:54) Yes,

Absolutely. It makes it easier for large capital to flow into the renewable markets.

Adam Sroka (26:04) So what are the key challenges from where you sit? Like looking at BESS projects, looking at solar, what are some of the key challenges, the really knotty problems that we’re having like generally? And then if you could share anything about any secret sauce about how Innova are innovating their way around them, I think that’d be really interesting.

Markus (26:24) I’ve just been at a conference this week and a similar question was asked. What is the biggest problem in the UK to deploy renewable energy projects and unanimously grid connection. If you go and want a grid connection, it could be in 15 years time. So that’s no good for anybody.

But saying that, what I’ve seen recently and I’m not 100% on top of it yet with the grid reform, where they want to go away from first come first served to a more developed approach where ready to build sites will be taken forward in the queue and make it more realistic to develop your projects. So I think overall, industry-wide, this is the biggest

hurdle to overcome. In next couple of years we will need investment in the grid infrastructure and we will need this clear process of knowing when you can plug into the grid.

For Innova, I think we are in a quite good position in that aspect. We have the grid connections.

But that is work of the past. How you work around that now, if you want to enter that stage is, and again, I’m not 100% on top of the new grid reform now, but if you develop your projects to a ready to build stage, you might be able to skip the queue a little and then get in front and get better connection dates. That involves the entire development process.

The further and the better your project is developed, the potentially better place in the queue you might get.

How to do that is a different question.

Adam Sroka (28:19) Very interested.

And what about like market and regulatory hurdles? Is there anything in that space that you’ve got strong opinions on or hot takes?

Markus (28:33) I think it really is regulatory, the grid. The market does the rest. So what the biggest problem in the market currently is, is the long lead times for, let’s say, transformers that take up to two years, which can also have a big impact on your project if you have that long breadth to be able to deliver it.

But the market usually regulates itself and especially with the big renewables push. Yeah, you need to work on your supply chain. You need to have strong partnerships, frameworks in place to be able to get your hands on that equipment in time. But that’s a market

topic that you can probably influence better than the regulatory side.

Adam Sroka (29:29) I think it’s interesting, like anecdotal, and I’d need to do a bit more research, but a lot of the people I’ve spoken to and the organisations that we work with really commit to very long-term, very deep partnerships with like one supplier for parts of the chain and kind of through thick and thin figure this out together. And it’s quite an interesting approach because it’s not something you see as often in some other industries. It’s quite like, and I guess it’s because some of this technology is so new.

But you’ve got to take the bet that, okay, that supplier is developing alongside me and we’re kind of building this chain as we go.

Markus (30:08) Yeah, I think going forward it will be more difficult to put all your eggs in one basket, especially because of what you said. If we look at BESS the technology kind of changes, I don’t know, every other month. And there is advancements here and advancements there and the containers grow in capacity. And we are now at, I think, six MW hours per container or something like that.

Which is a huge step. So what is it going to be in a year or two? Which technology is going to be the winning one? So you probably have to have a couple of partnerships to de-risk the technology advancements and be able to react and be flexible to get the best solution for your project in the end.

Framework agreements, if we talk at solar, usually with the panels, you want the tier one solar panels and then framework agreements to make a lot of sense there, which also gives stability in terms of if you commit to a certain amount of MWs with one supplier, you can maybe lock in a price to also help your business case with delivering the solar farms and have that continued supply that you need.

Adam Sroka (31:38) It’s such an interesting place.

The numbers of different roles and skill sets and professionals that you need to link all this up end to end, it often baffles me because you could spend 100 days speaking to 100 different people at any organisation like Innova and probably have 300 different conversations about what a day in the life looks like. It always fascinates me that companies can bring such a diverse team of people together.

Markus (31:53) Yeah.

Adam Sroka (32:13) And really make it happen. Like it’s really cool to see. I guess with the, yeah, the changing pace of things and just like interested, how do you find…

Like attracting the right talent and the right people and especially like data and techie people. Like is that easy for what you’re doing because it’s so exciting or is that a struggle?

Markus (32:38) I think it is a struggle. The entire industry is recruiting. So it is growing. It will be growing for the next years to come. So there is a fight for talent in the market, definitely. The data aspect is new. So as we’ve touched on it earlier, it wasn’t in the spotlight a couple of years ago in solar, especially, but it’s coming more and more in the spotlight and how to attract

Adam Sroka (32:44) Mm-hmm.

Markus (33:08) people that probably are more located in tech and working for big corporations to come to the sunny side and work for smaller but more maybe flexible and interesting companies in terms of having an impact on the power infrastructure.

We have a very, very good HR people department that are very active in that space. What I’m trying to do is to go into corporations with universities and working on research projects in terms of predictive maintenance and try to attract talent this way to give them an entrance into that industry by having a…

interesting project or maybe it’s a PhD or whatever we can offer to have first-hand experience in these assets and really develop something meaningful. That’s my approach and Innova’s approach in general to try to get the best talent here.

Adam Sroka (34:22) Yeah, like, yeah, the university kind of approach makes a lot of sense. There’s a lot of very clever people in academia that that want to make this transition to industry. And I think

Markus (34:30) Hmm.

Adam Sroka (34:33) the energy industry lends itself well to an easy landing for those people like some other industries don’t and it’s but yeah like people are excited I think people are starting to catch a whiff of how complicated and how fun energy can be and how techie and versatile and so it’ll be interesting to see how it plays out over the next 5 years because like you see all these curves and predictions of like battery assets going through the roof and renewables going through and you’re like

Adam Sroka (34:59) But who’s doing that work? Yeah, who’s the… what about the humans that are going to be involved in actually making that graph climb? I wanted to touch on before we before we kind of wrap up, I wanted to touch on your co-founder of a blog, aren’t you? Do want to talk to us about that?

Markus (35:08) Exactly,

Yes, yes,

My COVID project electrifying.world. Yes, when I came to the UK, that was in May 2019. And then COVID started in 2020, I think in March. So there was a lot of free time suddenly sitting in my flat in Hereford in UK, where there’s not much around it as well. So I…

Adam Sroka (35:38) Mm-hmm.

Markus (35:45) talked to a friend, we usually talked a lot about different sort of technologies and interested in that. And we just decided, yeah, okay, let’s make a blog. Let’s write down our thoughts. What we think about different sort of topics. And also maybe my biggest reason to do it was to develop my writing skills and then my English skills in general. I’m not that active anymore. I try every now and then to have a…

blog post coming out, but it was really intense back then in COVID where we’ve put a couple of posts together. And I will try to be more active again on the blog post to report about my learnings in the industry. And with that writing to go deeper into topics and learn myself.

Adam Sroka (36:35) Yeah, I totally, totally recommend that. As a writer blogger for years and years, I think it’s the best way to learn anything. And I always say like, if you ever find any technical content online that I wrote, I guarantee I wrote it as I was learning it, because I found that trying to translate what I’m doing into like notes and like a tutorial really sharpened my mind on stuff. And I think if you’re, for people out there looking for jobs and stuff, having a big public like presence and

Markus (36:56) Yes.

Adam Sroka (37:04) and sharing your thinking and work very publicly has done wonders for my career anyway, and I think it’s super valuable.

Okay, so look, fascinating conversation, guess any any final remarks or hot takes or any juicy insider stuff you want to share or stuff you want to say before we wrap up?

Markus (37:27) I think to wrap up the conversation around AI is take it easy, take it step by step, really spend some time in thinking about what you want to do. Put up a list of potential automations. If you can imagine it could be automated, it probably can, but then think about the process and getting it to paper and then into reality.

Will be extremely helpful going forward. I’m doing the same thing, constantly thinking about, or mainly the things that annoy me, the admin tasks, how can I automate them? And if you kind of develop that mindset and find a way to then finally get it automated, that’s a really big win.

Adam Sroka (38:20) Well, before we let anyone go, I always like to ask if our audience want to find out more about Innova or you, or if you’ve got anything you want to promote or plug, where should we point them?

Markus (38:35) Definitely have a look at innova.co.uk, our homepage. And if you want to get in touch with me, I’m on LinkedIn anytime. Markus Dirnbacher looking forward to your messages.

Adam Sroka (38:47) Well, absolute pleasure. Thank you so much, Markus. yeah, best of luck with all your sites coming online. You’re going to have a busy few years.

Markus (38:55) Thanks for having me, Adam.

Adam Sroka (38:59) I know I always say this, but that was such a great conversation I just had with Markus. It’s really interesting to see how actually far ahead they are in their thinking about data and AI compared to a lot of the people we talk to in the industry. When on the surface, like that’s not a core part of their messaging, but what you would maybe expect from someone that’s… or an organisation that’s developed 60 sites.

It’s almost like he’d parroted all the stuff that we always talk about at Hypercube. So it was a really fun conversation to have because I think we were really aligned and on the same page. They’re talking a lot about, do you trust the data you’re getting? Can you verify it? Like, how does that work? They’re looking at AI use cases and starting with small quick wins that are more probably leaning towards the admin side of the work, which is great

because everyone’s terrified of losing jobs. But if it’s like the admin bits that really bog down a job, that’s less scary. And I believe for where the technology is, that’s a really good place to play just now. And then iterating that to this kind of more advanced, more complex solution, they’re building a company-wide data warehouse and data lake and all that. Really interesting conversation. Fascinating to hear about the complexities of being an ops manager at a company like Innova

And yeah, just all round, really good guest. So thank you for listening and I look forward to many more positive conversations.