Julian: Hey everyone. Thank you so much for joining the Behind Company Lines podcast. Today we have Carlos Gaitan Ospina CEO and co-founder of Benchmark Labs which combines grid level weather data, site-based IOT sensors and propriety AI software to provide precise asset location located weather forecasting to optimize renewable energy generation and water allocation, reducing spatial basics, risk, and maximize financial benefits.

That was a mouthful. But Carlos, thank you so much for joining the show. I'm always excited, like I mentioned before the show, to talk to individuals who are working in the sustainability slash you know, data space within how we can better utilize our resources. Because there's always some interesting topics we dive into that are really, really fundamentally important for everyone in their everyday life.

So really excited to chat with you, but before we get into all that good stuff, what were you doing before you started Benchmark Labs?  

Carlos: Thank you. Hey, Julian, for the invitation, Being fan of Behind Company Lines. I think that it's a fantastic, actually name for the podcast. Thank you. So, yeah, that's awesome.

And yeah, to go to your question job I did many different things before Benchmark Labs. I worked at universities, work at the government in national labs, learning about how weather models. Were created contacting users, then were to then move to the private sector, working hardware companies, consulting companies.

And then it became clear that we had to create Benchmark Labs to solve a need that so many users were demanding a solution for.  

Julian: Yeah. And what was that need in particular that, that you were inspired to solve?  

Carlos: Thank you. Yeah. So many users were asking us, Hey Carlos, what can you do to help us to get better weather forecasts for our particular location?

It could be a farm, it could be a ski resort, it could be a really, a small area. While a traditional alternatives for where forecasting. They focus on providing the best available forecast for a region, let's say over North America or in California. Mm-hmm. and not they don't tailor the forecast to specific locations like a farm.

Julian: Yeah, well in particular goes into the forecasting. I, I we had a few, like I mentioned before, we had a few founders on who were, who were working within weather forecasting. And they, one was a satellite company. Another was more of an analytics as well. But what goes into the chain of events for us?

A, as a as an individual to know what, what, you know, our forecast is for the day. Like what, what is in the process, If you can describe how at the lowest level we can understand, you know, if it's gonna be sunny, if it's gonna be rainy, if it's gonna be cloudy you know, our localized forecast. What goes into that?

Carlos: Yeah. Thank you. Super insightful question. Cause in many ways a wear effect, all of us. and it's always a combination of multiple features of what is happening at your location versus what is happening. Also, it could be on the Pacific. Yeah. So to simplify the, the study a little bit, you need to understand.

Bigger systems that are going to appear or that will appear around your county, your state. And then you also need to understand what is happening at your specific location, which could be an area, for example, near a lake. Where you get more relative humidity because the lake evaporates all that water during certain times of the, of the day or near irrigation districts, for example.

Or you could be having different microclimates depending on the type of land cover that it's around you. For example you know, ski resorts have a very particular microclimates. Golf courses also are affected by micro climbs that they actually generate because in so many cases around the, especially here in Southern California, you see that the vegetation is so different inside of the golf courses and outside in the surrounding neighborhoods. So it's always a combination of both.  

Julian: Yeah. Well, when you say you mentioned the word systems a couple times. Are you describing like whether as a system and, and what's involved in, in the system that that we might not know and, and what are the different, I guess layers to, to that system or the different I guess variables that, that are incorporated into an overall weather system?

Carlos: Yeah, so so yeah, we consider the, the weather as a system. Basically it's a, it's a global phenomenon. It has like inputs, it has outputs. Yeah. It's governed by equations of motion. You know, we understand that the earth rotates. Then we have like fluids, like water in the ocean. We have at the atmosphere that is also a gas that moves very differently than the water and how the water and the air then you have also snow for example, on the mountains, creating a snow melt for rivers that will go into the oceans. Then you have ice cups, You have the north pole, you have Antarctica. You have changes in vegetation. That changes with seasons. And all of those ones affect the weather as a whole. So that's why we consider it kind of a system.  

Julian: Yeah, Yeah. Well, how can we better understand the system? Is it is it a combination of things? Is it having better instruments? Is it collecting better data, Analyzing data better? What, How can we understand our, our, you know, our forecasting of, of weather and data and allocation of resources?

Carlos: In many ways, it's all over the bob. You need better instrumentation, better processing tools, better cleaning tools of observations. So everybody contributes. It's more little grain of sand to the, to solving the problem. For example we have satellite data, as you mentioned that our founders came.

It's very important to understand. For example what is happening in the globe in areas where there are no observations, so you can get a better understanding of from a high level mm-hmm . But you also need a ground truth data where stations sensors that record what is actually happening.

And that could help to calibrate those satellite images because they are biased. They are drones, they are weather balloon. Aviation and then of course instrumentation in the ocean and in the depths of the ocean. So all of those ones are very important in many areas, to be totally honest. Are not instrumented at all.

The density of instruments is higher in the North Hemisphere as you can imagine. North America and Europe has like more observations of data than the developing. In the self atmosphere.  

Julian: So, yeah. And is the main result of not having, you know, instrument instrumentation just like being more susceptible to, you know, certain degrees of like, you know, magnitude if, if weather changes is, is it that, or is it something else that, you know, what, I guess what is the biggest fear?

For not having, you know, densely instrumented I guess sensors that can provide you insights on, on weather. Is it, you know, like I said, being cata, being susceptible to catastrophic events, or is it you know, just localized in terms of long term seasonality with, with farming and things like that?

What, yeah, what are the biggest issues with not having instruments in certain areas?  

Carlos: No, Fantastic question. The, the bigger problem is, If you don't have observations, you actually don't know what the truth is. Yes. So all your models say are inferring what? It might be a possible truth. They are you know, governed by questions of motion and physics, but there are many different possible solutions for the states that could be or could give, be given that will provide a possible weather realization.

A specific area. So if you don't have the real observations, the model will basically assume what is happening there. But if you have the observations, you can actually correct the model and say, Hey, no, actually you need more understanding of the processes. For example, you can say, The, the vegetation in this area is not high grass.

It could be low grass. And that by itself has a huge effect on how much water gets transmitted into the, into the atmosphere. For example similar situation with uh, a snow sometimes are parameters that you can tweak in the models that are possible. They have a range of, of possible values.

Some parameters, they call them parameterizations too values that for example could be possible between 0.9 and 0.95. But if you don't have observations, maybe a zero point 95 value, an example could be good enough. But if you have the observations that one, you will say like, Oh no, this is giving us.

Value that it will be too warm in a specific area and the errors will be too high. So let's tweak it. Maybe it's 0.94, you know, And that by itself, it's it's an improvement on the model. Yeah. This is totally oversimplification, but you get that point of . Yeah. Yeah. It's important to have like observations.

Julian: Yeah, No, a hundred percent. And, and I feel like the accuracy just, it, it has so many effects to different, you know, different people and, and different things. But I, I'm curious. One founder told me that, you know, or, and I even watched it in a, in a Netflix documentary and when it talked about weather and, and how, you know, the information is, is distributed but they were talking about how weather's becoming more commercialized.

And I'm not sure if you're seeing that yourself, but if you are, do you view that as a good thing or, or a bad thing? Or is it more complicated than, than good or bad.  

Carlos: Probably it's more complicated than good and bad. I don't think that necessarily has been more commercialized. Mm-hmm. , it, it I see that what is happening is that there are way more use cases mm-hmm.

Than what the National Work Services can provide accurate information for where technologies were derived. You know, by hand at the beginning based on observations, and it took like longer for the scientists to calculate the possible weather forecast than for, let's say the next hour. It took more than one hour to provide a forecast for the one hour.

So you know, that we have gone like great lengths since that time. But in many ways the, the weather models from National World Service. They were used for aviation national security, you know, even, you know, second World War, you can talk about invasion of Normandy. But there are many more use cases now, of course we have cell phones in our hand and everybody could just say, could have access to weather information.

So now there's like it's a very interesting time because there are more specific use cases, for example. Yeah. What happened? If I really need high frequency data, very precise of what is going to happen in the next minute, next few minutes, up to the next. You know, that's the domain of traders. For the, probably from the government, it's a little value because it's already providing a forecast that is good enough and serves 95% of the population.

But there is, this is specific use cases that what happen if I really need for information for. Just happening in the next 15 minutes. And that's important for energy, for you know stocks and trading. And so there's potential to commercialize in many areas where the current generation of weather models have gaps going beyond the, the, the weather range of 15 days, for example.

A huge amount of work being done trying to extend those capabilities and making it more precise improving the forecast for the middle of the ocean, for example. Yeah. So yeah, it, it, it could be commercialized, but of course it's because there is a gap between what is offered by the National Work services and what some specific users.

Julian: Yeah. Yeah. Well, in, in regards to like, you know the distribution of information how, how much does, you know, do governments in general control the distribution of that information? Or is it, do they control the access of it? Where, where is the involvement there and what is, what is your involvement?

You said you worked with government before and, and then you worked with the private sector as well. How involved are governments in the, the distribution of the information, the allocation, the the access? And is, is that, is it a positive experience overall?  

Carlos: Could argue that probably most of the governments in the world are heavily involved.

where forecasting here in the United States where forecasting goes under the umbrella of the Department of Commerce. But it depends, and each country is different. So we have here freedom of information. Data is paid by rate, payers by taxpayers. It has to be made accessible, but that's not the case globally in some areas, in some.

People have to pay for those services or the, the government doesn't. Those many governments charge for weather data. Yeah, if you want there, for example, what happened at the airport of, of Los Angeles? You want to get historical data cause you might be interested in a project or you might be interested in insurance.

Applic. Eh, some governments might charge for that level of data. Yes. Here in the United States you can get it for free. So it, it depends. And you can see that that opens the, the door for many private companies to, to go and fulfill those needs. And that's why, for example, IBM went in Acue, whether underground and they are way and. And service providers globally.  

Julian: Yeah. Well you, you mentioned you know, just having different access and, and involvement in different countries within, you know, weather and. What can we do? You know, we were talking about, you know, you're, you're originally from Columbia like you mentioned, and then, you know, I mentioned to you, my family's originally from Mexico.

How can we provide better, better kind of, whether it's data or, or weather forecasting in countries that are either, I wouldn't say, you know, countries like Columbia and New Mexico, they're, they're developed, you know, they're not like countries you know, in some countries in Africa, but there are some countries within South America that are underdeveloped as well.

But how can we offer better weather forecasting in areas that either. Don't have the instrumentation or don't have the resources to invest in that or, or maybe don't have the prioritization to invest in, you know, whether collecting and distribution of that information, how can, how can we provide better resources to those countries?

Carlos: Yeah, it's a fantastic question. There are many mechanisms even by governments, United Nations World Bank. There are grants, there are transfer mechanisms. Of technology. Yeah. So that's one way that the technologies get to be implemented in the developing world, let's say. Because also we have to take into consideration, as you mentioned, instrumentation, like the cost of an instrument here in the United States, if you earning dollars is X, but if you have to take into consideration exchange rate of Mexico or Columbia, or.

The evaluation. We even know what is happening with the supply chain and the economies globally, where even the British pound is losing its purchase power versus the American dollar. So you know, goods and services charging US dollars are in many ways less competitive now because of that difference in, in purchase power.

So that's where different organizations and mechanisms come. It could be subsidize the hardware costs to do technology transfer also from government to government. Those kind of mechanisms exist with the world methodological organization. Where scientists from the developing world. come to leading economies or government labs that have been doing like cutting edge weather forecasting for decades.

And then they get to experience and to transfer what they learned and bring it back to their countries. So yeah, there are many ways that, of course this can happen, but from the private sector perspective, we could or what we do, it's we try to. Offer solutions that could help them to be more competitive in, in the face of extreme events, in the face of managing labor saving water energy resources.

So in many ways some of the crops are commoditized and they are even. Grown in local currency, but they're paid in dollars, which it's not good for the locals that buy the product, but if the product is for export, then they get more money back when they translate it into their local currencies. So the cost of services, like the ones that we provide, for example, could be uh, in. Be transferred to the end user or the end customer that will be North America or Europe. Yeah,  

Julian: Yeah. And in regards to, you know, Benchmark Labs, tell us a little bit about the traction. You, you were you're Techstar back company, which is so exciting and you come with a big name behind you and, and you know, you're pushing this initiative to not only better the precision of weather data, but also to provide it, you know, to.

Maybe individuals or, or companies or institutions that, that had limited or maybe no access. What are some exciting feats that you've recently reached? Who, who are some exciting partners you've started working with and what are, what are some you know, some outcomes that you are particularly excited about as you continue to build on the, the recent success?

Carlos: Yeah. Thank you. Yeah. We're very excited about adoption of our technology since the time of Techstar. We now survey users in South America, Europe, and North America. We show that the technology can be transferred to the developing world and that it's a global solution, which is pretty exciting.

We had great testimonials. We have been recently in the news, even in ABC seven in LA where you are recording from. So we have a fantastic testimonial. About what they use, how we can help them to save water. And not only that, but how they. , their yields for their products were actually increased by having these high, high precise information to manage their, their crops.

So yeah, pretty, pretty exciting.  

Julian: Yeah, I mean, it's incredible to see how you like, like I, we mentioned earlier in the show is, is how the effects trickle down to the average. Person and really working together to kind of allocate our resources productively successfully. And, and, and with the mindset of, of long term sustainability it's just an overall benefit.

I don't think people can argue the opposite. Of that. What are some of the biggest risks that Benchmark faces today?  

Carlos: So we always have like of course internal risk and external risks. That can affect us, of course the funding environment to a certain degree. It's it it has an effect on startups all over the world in terms of like how we cease to get institutional investment.

So we want to scale. We'll have also to take into consideration what is happening at the macro level on the economy to try to use our. EH dollars and the contacts that we had earned, the money that we are returning in the best way possible. Taking into consideration if you know, what's the length of the financial crisis, what is going to happen next year is going to be worst or not.

So there are lots of macro forces there. That can affect us, of course, in terms of a runway that could affect how fast we scale, how many hires will be happening the next year to try to conserve our resources. But and of course at the macro level, you have like situations that are like, so.

They were stating like the war in Ukraine and that's affecting supply chain logistics the supply of Sunflower and that we probably, I dunno if you saw it in the news even that affected and created a very interesting problems in terms of all the sunflower, sunflower problems that you know, oils, butter.

Oh yeah. Many different prices have been going up and up and up up to 30%, 50% in some cases. Mm-hmm. and Ukraine has like such a big bread basket for Europe. So that affects everybody globally. So all of those macro situations you know, we have to take into consideration, for example.  

Julian: Yeah. Yeah. Well, if, if everything goes well, what's the long term vision for?  

Carlos: Oh, it's wow. We will be the go to where forecasting for specialty crops and high value land managers. So we are very focused on providing extra value for agricultural users and traditionally underserved markets. And for we're not focusing on wheat.

And so in can. We want to talk with growers that are affected by microclimates, for example, here in California, from s to be statues to grapes. Could be coffee, could be tea. So many products are vulnerable to wear but not necessarily have agricultural insurance or receive the benefits than rural crop farmers might receive in the Midwest.

So if things go well, we will. Eh, providing those forecasts to specialty crop growers globally.  

Julian: Yeah. Yeah. Is there any is there any crop in particular that you selfishly want to make sure always stays around? Anything that you, you like in particular? For me, for me it's like the pistachio one for me is I'd like that one to stick around

Carlos: That's fantastic. We'll, we'll try to see how we can help pistachio growers and yeah. Personally I would like to help of course growers of coffee. I'm from Columbia, and that's a huge part of the economy. Yeah. There and coffee grows and from, you know from Mexico to Central America, they are like millions of people that depend on high value coffees that are affected by climate change, by weather variability, and we can help them to be more competitive and that be a fantastic dream.  

Julian: Yeah, Yeah. Just kind of a sidebar question, what do you think In terms of, you know, industries that are gonna have, you know, more and more growth in the next few years is, is climate technology one of those that you see just growing exponentially as the years come and and technology around sustainability?

From, from my perspective is it's becoming so much of a you know, important. Reality that we all have to, to really focus on technology that's gonna sustain, you know, whether it's crops, whether it's, you know, the resources that we all share. But what's your forecast in terms of the growth of, of climate technology in your opinion in the next, you know, two to three to five years? Yeah. How do you see it now?  

Carlos: The the time is right for the clima tech companies, especially. Because there are so many opportunities with the governments at least even internally here in the United States and the European Union to a certain degree. There are lots of regulatory changes that favor Cleantech investments.

So while these economic Wins are favorable, then more venture capital money will come in the direction of Cleantech companies and investors will be more likely to take some risk in terms of development. Yeah. Yeah. So I think that it's a, it's a very greenfield to, to invest, to, to scale and to have a huge impact in the economy and in the living conditions.

Julian: Yeah. Do, do you have a a percentage in mind that, that you think it'll grow to in, in terms of where it is now? If not, that's okay. I always like to ask the question.  

Carlos: No, no. It's it is interesting, like, so yeah, it depends you know, instrumentation is growing at 17% year over year. There are new sectors that are totally new.

For example, fire. Yeah, so probably the, the order of like growth, it will be in the two digits easily for the next five to 10 years. Cause it's being held you know, underserved, underfunded but now people are getting aware of forest fire conditions, control burns what is happening, you know, like mitigation strategies.

And there's lot of technology that needs to be. And deployed if we want to be sustainable because every year now in North America, you hear about like a devastating forest fire. Yeah. And and you know, it has huge implications. Yeah. At the economy levels, at the personal level you know, generation of course.

And the livelihood of people living in those areas sometimes are counties that are decimated by forest fires. So It depends. And of course we have like clean tech in the traditional sense of probably renewables. Mm-hmm. Batteries storage that is very interesting. Alternative ways of energy.

So of course you have like solar, wind, but also there's potential for wave. Yeah, Coming from the ocean. Yeah. Different sources. So it's going to be.  

Julian: Yeah, no, it's exciting to see the different creative, like solutions that are coming out in, in so many different sectors that's really pushing this, this whole path forward to create a sustainable world and, and create or, or have resources that won't have an expiration date in.

And you're so right. I mean, in California, you know, if you go from one county to another, you might still be in a fire. And it doesn't just affect the, obviously it devastates the local community, but, you know, I think companies especially energy companies are becoming affected by it and are now seeing that, that it does affect them and influence you know, their business as much.

So it's, it's gonna be interesting to see where this. Space goes in the next few years. But before I let you go, I always like to ask my founders this question not only for selfish research, but also for my audience. What books or people have influenced you the most?  

Carlos: Works of people? How influenced me the most?

That's very interesting. I think that it comes for different periods of my life. I have like, they fortune to be around, not people that Say they are good influencer, like, Oh, that would be interesting. So for example you know, when I was in Colombia, I had the fortune to, to work with a very forward looking research group that the end only consider traditional engineering.

Practices, but was also taking into consideration the environment. So what's the cost of a, of a tree, for example, of or what's the cost of how you monetize or like, quantify the ability for your family to go to the lake and fish. Mm-hmm. So in those paradigms, I think that I owe that to my professor Nelson there in Colombia and the group that he created.

Then my PhD advisor. In British Columbia purely influential. He wrote the first book in terms of applications of machine learning in environmental sciences. So he, he helped me to lead the light of looking for applications and how to use these new technologies for the, the bigger goal.

Julian: Yeah, that's incredible. And you know, last little bit, I would like to give our founders a way to help us, the audience support your vision, your, your, your technology and, and where you're headed. Give us your LinkedIns, give us your website, your, your Twitter handles. Where can we support and find Benchmark Labs?

Carlos: Oh, thank you. Yeah. Please follow us benchmarklabs.com. Twitter is @LabsBenchmark and yeah, please. My personal Twitter is a @cfgaitan but we are happy to connect with the audience and help them in this entrepreneurial journey  

Julian: any way possible. Amazing. Well, thank you most so much, Carlos.

I hope you enjoyed your time and, and thank you as much for bringing on the show, giving us your insights from, you know, your, your career, your, your, your experience, your expertise and also interesting insights and, and things that we might not know about whether and technology and how it influences us, you know, on a macro and micro skills. So again, thank you so much for being on the show.  

Carlos: No, no. Thank you very much for the invitation. And yeah, please let me know how it can help in the future.  

Julian: Absolutely.  

Carlos: It was a great experience.

Julian: Yeah.

‍