Season 1 Episode 18: Robyn Lucas, Modo Energy, on the Data Science Challenge of the UK Energy Transition

This week, Adam is joined by Robyn Lucas, Director of Data Science and Research at Modo Energy. Modo offers a platform to deliver data and analytics for the grid-scale battery market and has a huge focus on improving access to information in the energy sector.

Adam and Robyn zoom in on the UK energy sector and the real-world challenges of data science, analytics, and modelling for this market. Robyn outlines the issues with data quality and the fact that, in this space, the future does not necessarily reflect the past. They dig into how regulatory and policy decisions affect the industry and why it is essential to maintain stability through any changes to ensure ongoing financial investment.

They share their perspectives on future trends, particularly in relation to energy storage, and touch on demand-side responses for domestic users. Robyn goes on to emphasise the importance of delivering simplified, explainable insights for industry stakeholders as it moves towards net-zero.

Robyn Lucas

In this episode, we covered:

  • Data quality issues that make modelling a particular challenge
  • Bringing more data science talent into the industry
  • The importance of stability in the sector for investment
  • Energy market reforms in the UK and their impact
  • Maintaining precision in data models without sacrificing usability

The weekly Hypercube podcast sits down with leaders in the energy and utilities sectors to explore how data analytics can help businesses make smarter decisions and accelerate business growth.


[0:50] Robyn gives an overview of Modo Energy and her background in the energy storage sector.

[3:30] Robyn and Adam discuss the challenges of data science in the energy sector and the difficulties of building accurate models in the sector.

[5:18] Robyn touches on the difficulties of finding and developing talent for data science roles in the energy industry and offers advice for students and those looking to get into the sector.

[8:07] Robyn unpacks the key takeaways from the second consultation of REMA and the considerations around pricing structure changes

[11:37] Here, the pair discuss the need for a stable investment environment and certainty around regulation to secure the finance needed for the green transition.

[14:01] Robyn discusses the UK’s CFD reforms and how they can be used to incentivise positive behaviours within the sector.

[16:55] Adam and Robyn talk through changes to the capacity market and the progress that is needed to support net-zero goals.

[20:27] Robyn predicts an increased demand for storage and flexibility in the future UK energy system, driven by wind and solar.

[24:07] Robyn gives a breakdown of Modo’s work and the different levels of focus on the areas of insights, modelling, and product development

[28:37] Balancing a full and accurate reflection of the complexity of the industry with accessibility and usability when building models.


Hypercube Podcast Transcript

Title: Robyn Lucas, MODO Energy, on the Data Science Challenge of the UK Energy Transition

Host: Adam Sroka
Guest: Robyn Lucas

Intro: Welcome to the Hypercube podcast, where we explore how companies in the energy and utility sector leverage data analytics to make smarter decisions and accelerate business growth. I’m Adam Sroka, founder of Hypercube, a strategic consultancy that supports asset owner-operators, traders, route-to-market providers, and energy services companies to unlock the power of data.

If you’re interested in hearing real-world examples of how data and AI are advancing the energy sector, this is the show for you. 

Adam Sroka: Hello, and welcome back to the Hypercube podcast. I’m joined today by Robyn Lucas of Modo Energy. Robyn is Director of Data Science and Research. For some members of the audience that might not know you or what Modo are about? Could you give a brief tour as to, yeah, like what, what’s Mo, what do Modo, what problem do you solve?

What, what are you about? 

Robyn Lucas: Um, so I guess, first of all, thank you so much for having me on the Hypercube podcast. Very lovely to be here. So yeah, my name is Robin and I, uh, head up our data science team at Modo. So Modo provides data and analytics for grid scale battery storage, primarily in GB and now in Texas.

So we’re a software business. We have a platform where we would like our users to come to get all of the information that they might need about past, present, and future revenues of battery storage here and in the U. S. And Modo’s been going for five years and we have quite a big kind of presence online.

We do a lot of videos. We do a lot of content marketing. So you may have seen some of our graphs and videos on LinkedIn. Um, and I have been at Modo for just over two years now. So previous to Modo, um, I’ve been in the kind of the sector with battery storage for about 10 years. So I worked at a company called Open Energy who was bought by BP a few years ago.

And so I headed up the data science team there. And we built a platform for operating flexible energy assets. So there was a lot of behind the metre stuff there. And then we did some utility scale battery storage as well. Things like hydrogen electrolyzers, to bitumen tanks, to water pumps, to aluminium smelters, and then to batteries, which it turns out, um, Much, much easier to operate owned by Shell, of course, for about six months.

And then I came over to Modo, one of the, I think, probably employee number nine or employee number 10, something like that. And now we’re about 35. It’s been an awesome journey, actually seeing very much in startup land and seeing the business grow. 

Adam Sroka: It’s actually nuts, like the surface area of the Modo externally, like super aware of it. I think if you’re in the energy sector, like it’s hard to miss the, the content’s great and it, it’s, it’s, I I think it’s really cool. So I could totally envision what it’s like on the other side of the fence and yeah, having worked in loads of startups and, and. Been there and sort of have some of the scars.

Um, so I guess then like, yeah, loads of experience in the sector and doing some of the techie stuff that like you’re doing now, but for other places for other assets. Keen to understand. Do you always want to know horror stories and stuff, right? So because it’s always funny, but do you have a thing that you thought would?

be easy, that was really hard, like anything that, that you doing Modo, you thought, Oh, okay, this would be a doddle. I’ve done this before. That actually was really difficult this time around and vice versa. Do you have anything that you thought would be hard? That would be quite, that actually turned out to be relatively straightforward and easy.

Robyn Lucas: I mean, there’s hundreds of things that you think are going to be easy and actually turn out to be really hard. If I think like back to before Modo days, you think integrating with like some really old pieces of kit. and then getting them to like turn on and off would be easy and actually that’s really hard and then you know fast forwarding today if you think about Maybe some of the things we do know is like building forecasts.

You want to predict how much demand is going to be in 10, 20, 30 years time. Uh, you know, just do some time series analysis of the history, you know, a few past couple of years, that would be really easy. Actually, it’s quite hard. I think that’s kind of like a lesson for data in the energy sector. It’s completely different.

I think to many other sectors where data science is much more established because the past does not look like the future and trying to train like big machine learning models to give you price forecasts or whatever you think will be easy coming into the sector and actually It’s quite difficult because you’ve really got to understand what the data is telling you and only take the bits that actually are relevant to the future, um, into your models.

I think I thought when I came in that I could get an imbalanced price forecast really easily. And actually it’s quite difficult. 

Adam Sroka: Yeah. Someone goes, I’ll just knock me up with a stack model. And you’re like, uh, sorry. Um, so that, that makes me think like, just to pull on that a little bit more, cause I’ve got ideas.

I’ve got thoughts. Do you think it’s because of other sectors? Maybe they don’t have the same, like, trifecta of, like, mass Actually, the hardware change is happening at a rate and there’s, there’s, like, the investment in technology is going on from a hardware perspective. And there’s a huge socio economic driver as well for everything to change.

You think that’s probably what leads to the data problem. 

Robyn Lucas: For me, the hardware thing is less of a problem. It’s getting a lot better. Like the data quality has got a lot better in the last five years. What is really tricky in terms of those like big machine learning models is that we sit on kind of the precipice of engineering, all of that hardware getting updated and changing its behaviours, politics has a massive impact and the economics.

And if you do a purely economics model, you’re going to get the wrong answer. If you purely look at politics, it’s very difficult to get masses of data from some decisions or press releases that have come out. And all of these things have quite a big impact and kind of the human element is, is actually huge.

And we don’t have as much data in the energy sector, um, as you’ll have in telecoms or banking. And the quality of it is generally a little, a lot worse. What you’re working on, you need to be able to pick up pretty quickly. 

Adam Sroka: And just to think about talent and hiring then, it’s the thing that I like to talk about a lot because it’s a really hard thing to do.

Finding a lot of people that want to do AI or do data in the energy sector. So, for example, like legacy energy. Generator or company or something like that, they want to make a move into doing more forecasting themselves or something like that, like try and bring themselves up to speed with all the hype and stuff that’s going on.

Um, they don’t know what to hire for and they like it, it’s hard to find the right people. And then the flip side, you’ve got grads and candidates that like, Maybe they are obsessed with the energy sector, but they just haven’t done anything yet. Like, do you think there are any good internships, projects, skills, and stuff that like a fledgling energy data nerd might go after?

Robyn Lucas: I think if you can, you know, there’s all those Kaggle competitions, um, which I mean, there are so many hackathons and, you know, exercises where National Grid or whoever else, UKPN have put out. A load of their own data that you can go and play around with and being able to come into the industry with some understanding of the language, because you’ve kind of had a little bit of exposure to that stuff is really helpful.

I think, and also like listening to podcasts. So your podcast, Modo podcast, there’s hundreds of others on the, uh, you know, if you go on to Spotify and have a look, there are, there are loads of others. Loads of really great podcasts that will help with that. Uh, kind of language and just kind of contextualising everything too.

Um, there are some really good masters, uh, courses that we do get quite a few candidates coming in from. Um, but really you want to try and get as much hands on experience as possible and, Those master’s courses in data science don’t always provide that kind of really hands on nitty gritty stuff that you want to actually be going and scraping the data and trying to get something decent out of it yourself.

Adam Sroka: Yeah, no, I totally agree. I think it’s a mixed bag that some of the courses are like four or five years behind or they’re trying to give too broad an exposure to like every tool that you could possibly use. Whereas in reality, like you, you wouldn’t do computer vision one day, NLP the next, time series the next.

Like that’s, that’s just not the way your career will play out usually. 

Robyn Lucas: Exactly. And like a load of image visualisation stuff is totally useless in this sector. 

Adam Sroka: Get into the energy sector stuff a bit then. So we had the second concentration of Rima. Um, this year, you explored this and a few other related topics on transmission.

I’d love to kind of hear some of your thoughts. Things around, you touched on nodal pricing in Texas compared to over here. What are your thoughts on that front? 

Robyn Lucas: Yeah, so the second consultation of RIMA, The consultation closed, I should know the date, but like a month ago, and there are a number of different parts to it.

So one of them, one of the big news out of the first consultation was, is the UK going to move from national pricing to zonal pricing to nodal pricing? So this is where the wholesale price of electricity changes depending on where you are in the country and zones there might be like, well, depending on who you talk to, 2, zones.

Whereas nodes will be much more granular, maybe every grid supply point. So I think that’s like 360 of those in the UK. Potentially there would be a different power price in each one of those. I’m going to discuss the ORP, the Australian or the UK, uh, Transcription Provided by Transcription Outsourcing.

org. so much for joining us. I’m so glad to be here. So, uh, I, uh, I’ve been working in the, um, And that means that some of the boundaries that separate different regions of the UK get really constrained and that leads to costs in balancing those constraints. So namely here, we’re talking about onshore wind north of the B6 in Scotland costs us like tens, if not more, millions of pounds every year to turn down and then turn things up south of the border because that boundary can’t cope with it.

So, um, a local price is a much more efficient way potentially of dispatching everything, um, kind of in merit order and dealing with those constraints rather than turning things off and on and on again in different places. Um, so in the second. consultation of RIMA, uh, nodal pricing was taken off the table and we are just left now with zonal or maintaining a national electricity price.

Um, and I think there was quite a lot of pushback from the industry regarding a nodal price. Um, it’s a huge change from where we are now. And there are all sorts of issues around this kind of liquidity. Um, you know, what if you’re in a node which has loads of wind and no demand? Are the assets that you’ve got there then going to make no money?

And, you know, how do you deal with grandfathering? There’s a lot of questions, um, about how it would be put into practice in reality, given how well, um, Or not so well that National Grid are able to transform their, uh, IT systems. Um, so yeah, we are left with zonal and national and uh, I think the Modo opinion is probably that we’d quite like zonal because there’s more data involved. Um, but I think we might stay with a national price. 

Adam Sroka: It’s such a hard change to bring about, right? Because I’ve got Yeah, like for all the reasons you’ve mentioned, but you think even like. Got the capital expenditure to build some of these wind farms and things like that that happened maybe 15 years ago, didn’t factor in a change in pricing structure and so on and so forth.

So unless it definitely benefits whoever’s spent the hundreds of millions to build the thing. That’s, that’s like you’ve banjaxed a business model potentially, or you’ve really upset the apple cart. 

Robyn Lucas: Yeah. And that has massive implications for future financing. And we need to build, what’s the labour manifesto saying?

50 gig solar by 2030. And the Tory one isn’t that far behind it, 2035. That requires a huge amount of capital and You know, the UK has done very, very well in terms of green investment in the last sort of five, 10 years. We’ve managed to greenify our grid hugely, right? Much, much better than lots of other countries because it’s been a relatively stable investment environment.

Things are a little bit more tricky now with higher interest rates, but if you’ve got this huge regulatory uncertainty of what’s going to happen In terms of like, you know, your bread and butter of your business model, which is your merchant, your merchant revenues, which will underpin everything else that creates a huge amount of uncertainty and people will go elsewhere to other geographies where you have massive tax breaks for investing in similar sectors.

So yeah, it’s, it’s, it can be quite dangerous, really. 

Adam Sroka: I’m not going to make a comment either way, but it’s just that it’s hard because what’s right for the green agenda and whatever really just depends on the timeliness of what you’re looking at, like, do you want to make it right now? Or do you want to incur the pain of that transition now?

Do you want to just make sure we get the right level of investment? Yeah, I don’t envy those decision makers. 

Robyn Lucas: Yeah, it’s really, it’s really tricky. No. And they’ve got to, you know, take on board so many different opinions when making the decisions as a result of what comes in as in the consultations. Um, but yeah, that’s, that’s kind of.

national and zonal pricing, but there are lots of other bits of RIMA that potentially impact storage and kind of the wider industry. And it’s like, you can’t really treat one of those things in isolation. It’s quite a holistic approach that RIMA is taking. 

Adam Sroka: What about the CFD reforms? They came up on the podcast.

Robyn Lucas: Um, the CFD, so it stands for contracts for difference. And this is how most. Uh, wind and solar has been financed. Well, wind for sure, maybe solar less so. And it kind of makes these projects investable because it provides a secure long term revenue stream. So a CFD contract is like 15 years long and it essentially works as if giving you a fixed price for every megawatt hour of power you generate, which works really well when you’ve got a fairly small amount of renewable energy on the system.

So maybe I know 10, 20, 30 percent of your capacity is safe from wind and solar. But when you start to get up to kind of 30, 40, 50 gigawatts, which might be, I don’t know, 70, 80 percent of your overall generation at some times of the year. That has huge effects on the wholesale price in those times, because suddenly you’ve got all of this energy being dumped on the system, which means that your wholesale price, you know, it’s supply and demand, your wholesale price crashes, and therefore you get price cannibalization of those assets.

So if your CFD contract. which is guaranteeing that project a particular pound per megawatt hour, um, regardless of what they generate and when they generate. That becomes very expensive for the taxpayer, and it doesn’t incentivize the behaviours that we would necessarily want, because if there’s lots and lots of wind on the system, we might not need much more wind, um, and actually we want to turn down wind and not pay to do it.

And we see this happening, you know, Okay. Right now we get negative pricing in the wholesale market and wind is not being incentivized to turn off and therefore we end up paying it to turn off, which is very expensive. Whereas if we were to reform that CFD, then we might make that whole situation more efficient while still maintaining really positive investment signals and the secure revenue streams that people need in order to, you know, spend the millions of pounds they need to to develop those projects.

So one of the options on the table is a deemed CFD, which is much more about incentivizing the right behaviours. So you would get a payment based on your capacity over a year or your availability rather than for every megawatt hour that you generate. 

Adam Sroka: Not straightforward and it does make you think like, how quickly would these reforms be put into place?

Like how, someone knows it’s always going to be out of joint. 

Robyn Lucas: Yeah, I think I said on the transmission podcast, if, if everybody’s a little bit annoyed about all these market changes, you know, the effect of them, then they’re probably doing quite well. It’s about compromise, right? In terms of timelines, um, I think they would like everything to kind of be done and dusted within the next five years.

So it’s not, it’s not immediate, but for sure within some of the kind of investment timelines that people are looking at, particularly if they’ve got grid connections in like three, four, five, six years time, um, it’s definitely something that they should be thinking about. 

Adam Sroka: What about, uh, capacity markets and some of the reforms there?

Do you think they’re enough like to get to where we need to get to for the net zero. 

Robyn Lucas: So the, um, the capacity market is how we, it basically acts like a subsidy and it’s how we, um, build the supply stack is how we decide what the generation mix is. And, um, we, at the moment, there are derating factors for different kinds of technologies.

So. These are particularly bad for batteries because batteries are duration limited technologies. So they have quite high derating factors and they’ve gotten higher and higher over the years. Whereas if you’ve got like the CCGT are more like a hundred percent because you’re always online and you can kind of always keep producing power at times of system stress.

So the quality market, I mean, I guess the big thing is it’s a very expensive market, right? The contracts that we handed out this year are worth billions, billions of pounds, 60 billion pounds, something like that. I could be completely wrong there, apologies. But you know, an awful lot of money. And we had the highest ever claiming price, which was like 70 pounds in the team for us, and we got no new built gas generation.

So the capacity market was originally designed to incentivize new built gas because it’s a secure supply of power. clean and, um, the government at the time were very pro gas. I think the government is still very pro-gas if you think about the press releases that were put out at the, around the similar time of, uh, the RIMA consultation earlier this year.

And gas certainly has a part to play in helping us get to net zero, but currently you could get a capacity market contract. With a, you know, quite badly emitting gas plant, which would give you money far beyond when we’re supposed to be running a net zero system. So there are some changes to the capacity market, which will have different pots of money.

So you might get a, uh, they’re called minima. So you might design some bit of the system that you want to, uh, procure capacity for. So that could be like, I want low carbon, long duration flexibility, or I want some very short duration stuff, which is going to deal with system stress events that only last a couple of hours.

And there will be technologies which will be able to handle what those minima are requiring and each one of those will have a different price. And there’s some other options in there, but kind of do the same thing, basically trying to reform the capacity markets so that it is actually going to build the generation stack that we need to get to net zero.

Adam Sroka: Yeah, like on the gas stuff, a lot of the existing kit that’s out there is pretty old and like incentivizing it to sort of stay on and things like that. People aren’t like, there’s not, people aren’t excited to be investing in new gas just because of, you know, The wider picture, like the, again, the socioeconomic trend, the attitude culturally towards it.

Robyn Lucas: Yeah, exactly. Which you can see in the auction results, right? There were two gas plants, which were in the T minus one auction. So that’s for this coming winter securing capacity. And they came out of the auction before it closed because even with really high clearing prices, it wasn’t enough to keep those plants alive.

So they’ve got such high operating costs now. And there’s, you know, uncertainty about longer term revenue generation, depending on what the outcomes of all of these changes to the markets will be. And, you know, with much more wind on the system, lower wholesale prices, generally it’s a very uncertain business model, which is kind of at odds with what some of those government press releases have been saying regarding, you know, is very much part of the future.

Adam Sroka: So any kind of Importance or trends for the future. Do you think they’re going to make a big difference in, like, what, what’s your kind of big bet for the most positive change over the next sort of decade? 

Robyn Lucas: I think we are going to have, we’re going to have the need now and in the next 10 years and beyond for lots, lots more storage coming onto the system.

We’ve had, the last winter was fairly flat wholesale prices, of course, coming off of two years of incredibly volatile wholesale prices, um, and ancillary markets are saturated now and have been for quite some time, frequency response revenues, which is like the mainstay of how batteries make their money, that’s kind of very much gone away.

And most of the money is coming through, um, the merchant model. So this is through wholesale or the balancing mechanism. So we hear a lot about it. You know, the market for storage is cannibalised in the UK and I really don’t think that is the case. It’s cannibalised in frequency response services, but that was always a really small part of the, you know, very short termist view we’re going to have so much more wind coming online and solar coming online and we already see so many like zero or negative pricing, that’s going to be a massive thing in future and storage is really the only thing that’s like a part of that.

Um, and along with the BM, so that changes that. Uh, coming through the control room that we’re already seeing the effects of that storage is going to get much more money essentially through the BM. Um, and actually be able to kind of use its flexibility in the best way for the system. It’s perfectly suited for the balancing mechanism.

Yeah, I think we’re going to see a lot more storage being built than some of the big names in the sector might suggest. 

Adam Sroka: You see some interesting stats on that and people banding around, like comparisons to barrels of oil and stuff per battery and things. And I think, um, With more residential movement as well, like the kind of residential holistic home energy management stuff, I think we might see that boiling of the ocean, like the temperatures of people’s experience and understanding rise and the kind of the slow.

Movement over to more of that in the collective consciousness and it not being something that’s like brand new and exciting. And I think, uh, going to some of the events, you see the amount of people there that are excited about storage and new technologies and that kind of thing. There’s definitely rumblings going on in that part of the industry.

Robyn Lucas: Yeah. The domestic side with some kind of really novel supply contracts, um, you know, some of the big names in the sector are doing a huge amount. Yeah. On demand side response. So yeah, really interested to see how that all plays out. I think it’s. It’s more difficult than people might think. Um, but yeah, it’s, it’s, it’s very exciting.

And then like the best thing about this industry is it changes all the time. 

Adam Sroka: And more so than other industries I’ve worked in, like everyone cares, like everyone cares. Like you work in retail or pharma or manufacturing. Some people care, but most people don’t. Everyone cares about energy and they can get it.

So it’s quite cool. Modo put out a lot of, Really good, like new novel market analysis. Loads of really sexy graphs about stuff people hadn’t seen on LinkedIn, and they look really good. And so what split of your work is like long time or broad analysis work, like interesting insight generation versus like crunchy model building and like data science y stuff?

Robyn Lucas: Yeah, this is definitely something that has changed as the business has kind of matured. So we have a data science team, and then we have a content team and we have analysts in both of those teams. So Wendell heads up are the stuff that we put out externally. So all those graphs on LinkedIn, he has made the most beautiful graphs and that team really focused on like the here and the now, so those are the faces that you see on videos and then, uh, my team will take those results and then incorporate those into our models.

To kind of productize them a little bit more. So we can churn out articles and videos and graphs very, very soon after the event. To put that into code so that it works and then put it onto our platform takes a little bit longer. And if we were to do it with everything, we’d have a really disjointed massive product that, you know, Would probably be quite difficult to use.

So there is a whole load of distilling and processing, which goes on, where we decide what we actually want to put on the platform and also what goes into our models for either benchmarking recent, uh, revenues or what will go into our forecasts. into the future. So I guess a good example of this earlier today, I saw that Zach had put out a really nice graph of like wholesale prices versus DR high.

So dynamic regulation, high clearing prices, and there’s a really nice inverse correlation. So we would use that to be able to inform our modelling of what we think DR high prices will do going forward. So that way our forecasts, um, are kind of very reflective of today’s markets. And we feel like we have a You know, a really good insight into how we think those things are going to change and Implemented like the latest observations we make through our market kind of facing team.

Adam Sroka: It’s always like early warning signs and like an extra kind of research wider view thing that’s like, not. Gives you a chance to kind of filter some through, I guess it leads into something that not a lot of people think about in, in data science and modelling land, the importance of like a product owner or someone with a vision for the product, like you got any views on that or opinions.

Robyn Lucas: Yeah. So we have a product team at Modo who spend a lot of time talking to our customers. So what I think is interesting is probably not what other people would think of interesting. And if we, and they provide a really good kind of litmus. And getting a load of user opinions into actually what we should put on the platform, because if we were to do everything that we thought was interesting, we’d have this huge product probably that was really disjointed, that would be difficult to use and wouldn’t have like a really clear sense of design or like the UX side of it would be quite poor.

So the product is very important and kind of actually what we present to the world as well, you know, a really coherent view that we have. Um, polished and filtered to make it as easy for you to understand rather than, you know, the ramblings that go on in mine or any one of our heads. 

Adam Sroka: On the, uh, the models themselves, what’s the kind of balance and like, I guess, because this might come from your product team, but what’s the balance between like the most accurate, the most precise models, best like AUC or whatever versus Being able to understand it and like actually explain the outcomes.

Robyn Lucas: It’s, it’s a really tricky one because you need to have enough faith in the numbers that your model is giving you and that you can understand the outputs of that model, but you have to be able to explain it. And you have to be able to tell a story with that model. So we, we spend a lot of time thinking, is this good enough?

So the way that we work is to try and do an MVP. We’ll try and do something simple and we will always go for simplicity rather than complexity because it has that explainability, um, and everybody can understand it and therefore poke holes in it as well, which is really great. And we will say, is this good enough?

And if we can’t back those results up with, you know, actually wholeheartedly believing them, then we will go back and go a little bit deeper and a little bit deeper. But we try to very much stay on the side of simplicity and explainability. So I will always use a linear fit as opposed to some longer machine learning model, because if it’s good enough and good enough might be like, I don’t know, 80 or 90 percent of the way there.

We are in a commercial business, not a university. So we need to get results quickly. But yeah, we have to be able to stand behind those results as much as possible. 

Adam Sroka: Couple of points. So where, where do you see data science and like general research into space? How do you see that evolving over the next few years?

And like, what advice would you give people that are just starting this journey? 

Robyn Lucas: I think, um, AI is going to be massive, right? And a lot of, maybe more of the technical skills that you might have developed a few years ago. You don’t necessarily need to do that anymore because AI can do it for you. So. What I think it will struggle to do is really understand the context of the industry in which you work.

So as a data scientist coming in, I would really recommend choosing an industry that you are passionate about and learning as much as you can about that industry, and then trying to figure out how your skills can help solve problems that need solving and it could be making things more efficient, making things quicker and, you know, where your intellectual curiosity is going to get satisfied.

Actually, where you’re really passionate about what you’re doing as well, because that’s, that’s one of the things I love most about my job is I’m super passionate about it. And everyone I work with is super passionate about what we do. We talked about this earlier. Um, and I think that’s really important for anyone coming into the sector.

Adam Sroka: I think I’ve got blog posts after blog posts written on this, but totally the everywhere I’ve ever worked the best data scientists are those that have great domain knowledge and you only really get that by just being like a little bit obsessed and nerdy about the thing you’re trying to solve. Like we’ve all worked on projects where We don’t actually give two hoots about things like the widget or whatever that’s being optimised because it’s just an exercise.

But when it says your thing, lights go on, like the colours change, it’s all amazing and you can’t run out of ideas. Robin, it’s been a pleasure. Before I let my guests go, I always just like to ask a similar question, anything you’d like to plug or promote, or if people want to find out a bit more about what you do or what Modo are up to, is there anywhere we can point them?

Robyn Lucas: Yeah, so Modo put out a Tons of content on LinkedIn is probably the primary place for that. Um, so do go check out our page. So we have a UK page on a USA page, um, for all of the fun graphs about the latest in battery storage revenues. Also go check out our platform. We give away masses for free. So lots of content and you can make some really great plots and see what we’re up to on the 

I’d also say if you want to learn a little bit more about our forecasting and our API, you can get those links from the platform, but we, We have kind of completely transparent, open methodology that anyone could go and read about, and then, you know, query, ask any questions that you want, and we love to hear from you. So, please do get in touch. 

Adam Sroka: No, look, can testify to that. They’re really well put together, the kind of platform docs and the way that’s all presented. And I force feed most of our staff, like the Energy Academy, YouTube and, and things like that. It’s just really useful. It’s a lovely, succinct way to consume content that isn’t like a dry book or a white paper or whatever so.

Robyn Lucas: bite sized videos that actually are interesting and entertaining is really where we’re trying to go. So, uh, yeah, check it out if you’re interested. 

Adam Sroka: Great. Well, look, it’s been a pleasure. Thank you very much for joining us, Robin, and, uh, wish you all the best at Mago. 

Robyn Lucas: Thank you so much. Thanks for the chat, Adam.

Outro: And that’s it for this episode of the Hypercube podcast. Thanks for tuning in today.

If you have any questions about the topics we covered, you can reach out to us on LinkedIn or check out our website at You can also join Beyond Energy, our Slack community of data leaders from the sector. There’s a link to sign up in the episode description. We’re just getting this show off the ground, so if you like today’s episode, please leave us a rating, review, or subscribe wherever you get your podcasts.

It all really helps. See you next time.