AI for Utilities with Vik Chaudhry, Co-Founder & CTO @ Buzz Solutions

Oh, today on the show we have Vic Shawrey. He's the co-founder of Buzz Solutions. Buzz uses uh AI and machine learning, but not in the buzzwordy way. Uh they legitimately have built a technology that very positively impacts the ability of utilities to manage their assets. Now we're talking about forested overgrowth, you know, like trees uh

getting in uh electric lines, talking about other wildlife uh that could be endangering themselves by coming too close um to power lines. And uh all told this uh this management uh is extremely uh behind its time as far as the technology applied uh to manage the assets. And all of this is around um as

the demand for more grid capacity grows um every outage means more and more. So uh very incredible story. Uh they've been eight years. Uh they've been commercialized for about half of that. Um it's very interesting exploring his headsp space uh navigating that chasm between research and commercialization uh his background uh in the technology

and navigating uh the culture at Buzz that supports them uh continuing to grow. So shout out to Clean Tech Growth Lab. Uh if you're growing in clean tech, you want to work with them. And uh shout out to Craze and Friends for producing this podcast. Uh, and now I give you Vic.

Oh, welcome to another episode of the grow. Thank you to the partners. Shout out just before we pressed record here, but without them, it would not be possible to interview awesome people doing awesome things like Vic. Welcome. Thank you, Blake. Great to be here.

This is so cool. Uh, I don't think I want to give I was already asking you a lot of questions before we started. uh recording just because there's so much um uh that you have to offer. So uh why don't you just start with uh giving a brief introduction of yourself for anyone that doesn't know yet.

Yeah, certainly. So I'm Vic Shadri, co-founder and CTO at Buzz Solutions. Uh I come from a background in technology space. So prior to Buzz, I was uh leading MLA AI teams at Cisco for a few years and then got back to my energy roots. By trade, I do power systems engineering, energy engineering. That's

what I pursued at Stanford University doing my post-graduate degree and wanted to apply cutting edge technologies like data analytics, data science, machine learning, AI to that sector and this is all before the whole AI um hype and AI boom as well with with all these large language models. So I've been practicing AI for the last 12 13 years working with

various kind of tools and then applying that to the energy sector specifically in demand response deers uh you know smart grid technologies those kind of use cases and that's where I stumbled upon uh Buzz decided to start this company with my co-founder uh out of Stanford University and then we've been doing this for the last 8 years now

working with power utilities power companies helping them you know make their operations safer efficient with with our AI techn technology. That's right. So, uh before we get into again I've said I've already said it I think twice in like the five minutes I've been on this call I think uh you know eight years uh committing to

something like that is uh for that long uh is rare. You can look at the statistics and I think it's u uh much respect for doing it. So I'm excited to talk about that journey. Uh before we get there, for you personally, were you walking around crawling around as a baby saying, "I can't wait to work on AIML,"

or was there some point or something where uh this technology um was interesting enough to you uh to pursue it? Yeah. No, great question. I actually I come from New Delhi, India. I uh immigrated over here in 2015 um to go to grad school and then pursue uh opportunities over here. So growing up

in India, power was a big big problem. So we used to have 5 to 6 hours of power outages every single day. So growing up facing those kind of unreliable power systems got me thinking that there must be some way of helping power companies make electricity which has been invented for 100 years now

uh be more kind of outreach to people. So so that got my attention early on that hey energy is is a big area where that needs help and that's where it needs technology infusion. But then growing up in New Delhi um there was a big problem of pollutions specifically air pollution. So when I was in my

undergrad we we used um well I built with my team a drone a drone from scratch and flew it around the city of New Delhi with some gas sensors um integrated on that to collect pollution monitoring readings and we were able to find that the pollution was so bad that it was comparable to smoking 20

cigarettes at the same time. So, so that that got me thinking that, you know, at the same time, the same time it was it was pretty bad. Uh, but that was kind of the start of my journey with data. So, because we were collecting data from sensors, we were able to kind of predict also cuz we were

collecting historical data. And that got my uh thinking into hey there is a future where data and data analytics and data science and machine learning can be applied to critical sectors which was missing at that time. This is 2012 2013 timeline. So that's when I got kind of interested in how innovative technologies like AI, ML, data science

can be used for real world problems. in my case it was energy and that was kind of the start of my journey exploring data science and AI and how we can apply that to real world problems. So, so it was so it was the the drone project and then all the data that you were able to

collect with that uh you were in an environment where people said hey a way to digest this data and get something out of it there are technologies like AI and ML and that's how you got introduced. That's correct. Yes. We started doing some probabilistic modeling and that was my kind of into the AI world

and then you said wow I love this. Yeah it's so much you can do now. So it it seems like a uh it seems like a big opportunity um seems like a really big decision uh from my perspective to pursue uh grad school you know across the world. So what you know was it kind

of a natural next step for you or was that a big leap for you personally to um to to pursue school that's so far away from home? Yeah, it was definitely a big leap. Uh came here with knowing no one. uh wanted to build a network and pursue opportunities over here. Um and one of

the key things was getting an invite from or acceptance from Stanford University because heart of the Silicon Valley that's where you know the cradle of entrepreneurship originated you know so many founders there's so much innovation happening at the heart of Silicon Valley um at that time and now it's there's there's hubs everywhere

which is great we want to see more innovation but I wanted to be in the technology center of the world and pursue opportunities and apply technology to some real world problems and and not just you know optimizing or generating ads for for platforms. So that was that was the key and then at

that same time I looked at Stanford's program in energy and they were doing some great work. So there's a good network of people that are working on really cool projects in energy. So that got me fascinated. I had a job lined up uh in India and I I I ditched that and then decided to pursue grad school come

move to another country with no connections and then took a bet at that time. Maybe I was uh naive as I was young but you know hopefully that worked out well I I'm glad I know there's tons of people over in Silicon Valley that are glad uh as well and that works.

So, so then so then there there was a period of time between uh your time in grad school and when you decided to start the company is that right? That's correct. Yes. So at grad school I was still pursuing a lot of projects in energy uh you know specifically in uh deers you know renewables was a big kind

of use case for us in grad school we were working on energy problems power system problems uh but also I got involved in other in another drone project in my research lab where we were using drones to optimize wind turbine uh locations and kind of see how the condition of wind turbines are. Um so

that was kind of again trying to use drones to do some kind of data collection but then use data science machine learning AI on top of that. You were you were using the drones to see where uh high potential areas for for for wind turbines not like flying around check on existing ones.

That's correct. Yes. So there were certain category of wind turbines we were kind of exploring which was vertical axis wind turbines that are smaller than your conventional wind turbines but they rotate on vertical axis and and they are and that was kind of a new kind of technology and we were trying to figure out uh obviously with

drones and wind sensors on it where are high potential winds that can help optimize the locations of those. So yeah, cool cool cool projects with drones we're pursuing and then that kind of got me into an area where Buzz Solution launched in, you know, summer of 2017. I met my co-founder over there,

discussed some ideas on how we can use AI and and apply that to the power sector. But there there there was a there was a job at at Cisco before then or during before that. So yeah. Okay. So, so some something that's uh very fascinating to me is that threshold of hey, I have a life, I have a job, you

know, whether it's school or um you know, a corporate uh like some something more steady uh and the and the decision to pursue an idea that you feel passionately about or you know starting a company like that. So, can we talk about where the the idea? So, what was the initial before we talk about how

it's changed? What was the initial idea of Buzz? Um, and where did that come from? Yeah. So, so we both me and my co-founder were at uh Stanford and we were in a launchpad course that fosters entrepreneurship in uh real world problems. So, that was a great course we took. That's where we got connected um

and discussed the idea. So initially uh the research that was happening in my research lab with the drones for win turbines that was kind of the idea in that launchpad course but then both me and my co-founder naively started cold calling power companies which is you know again we didn't know the industry that well but it kind of

worked out. uh we we got in contact with 30 different utilities discussed that idea with them and they said why don't you apply this technology for the power sector and that got us thinking and then we decided hey it's a much bigger market they have much bigger problems inspecting power lines they utilities

were already starting to do inspections of power lines with drones uh so initial you know first couple of months we were also going to fly drones and then we had a software to analyze data but soon we realized that hey none of of us are drone pilots. B, we don't want to be in

a very saturated drone market. And C, we don't want to bear the liability. What if a drone crashes into into the line? Sure. So, we decided that we're going to be a full software play SAS platform uh that analyzes data captured through either drones or helicopters or any kind of aerial vehicles uh around power lines or

any kind of power infrastructure and then make sense out of it for power companies. So, so that so that was the idea from the beginning. That's correct. And and so did you again just clarifying the timeline. So you guys came up with this idea and then you went and and had a job for several years

and then launched the company or the job came before that? The job came before that. So interestingly you know launched the company we were working on the product. So you know we've been in the uh we launched the company it's been 8 years since incorporation but for the first few years we were building the product.

So from the timeline of 2017 till 2019, we were building the product. We were building these proprietary AI models and getting access to a lot of proprietary data from utilities and and different partners. Full-time or part-time building the the product uh full-time building the product.

Wow. Okay. Yeah. And and pursuing that. And then after that we started commercializing in 2019. So running proof of concepts with utilities and we didn't we started running pilots and then COVID hit. So that was an interesting time as well in 2020.

Yeah. Interesting. Yeah. But but the good thing was you people needed power and utilities are recession proof industry. So utility it was even more demanding for utilities to kind of use remote sensing technology like us. Uh so we got in contact with a lot of utilities. We raised our financing round our seed round of

funding during 2020 and that's when we started scaling. So a every the scaling started happening after 2020. before that was all product development. So then so then what was there in those two three uh I guess even well let's say two to three years before you properly commercialized was there consistent momentum towards commercialization or

were there times where you said this doesn't make any sense I don't know what we're doing anymore you know there was momentum but as the journey of entrepreneur there's always ups and downs right there there's sometimes doubts can arise in your mind But one of the key things, one of our mentors told us that one of the key

skills for an entrepreneur to have is persistence and patience cuz it's always a buildup. Things don't happen overnight. You have to put in the work today to bear the fruits and the positive outcomes for the future. So we were very kind of one one of the things that some of our customers and partners

told us is that you guys are persistent. So one key thing that kept us going uh because again both of us were young maybe we were naive but it worked out. We were very very persistent. So in that time period also we went through a bunch of you know accelerators uh got in build

build some relationships cuz energy industry is very close-knit and we were coming as outsiders. We'd never worked at a utility. So we were able to build some relationships set some reputation as well build partnerships in that time frame. So there was consistent momentum in that direction. So then so then what was the what was the another threshold?

What was the threshold of the first uh commercial deployment? How did that go? Yeah, that's and that's always interesting. Utilities move slow because they're riskaverse uh to new technologies and and fairly they they should be because there's a lot of new technologies and that might not work out uh for utilities and if they don't work

out you know people lose power, people lose lives, those kind of things. So what we went through as a first deployment with a real customer uh was through an RFP process. So New York Power Authority was one of our first customers, first enterprise customers. They put out an RFP uh request for proposal bid in 2021 and we

went through that and you know over 100 vendors applied for it with the like likes of heavyweights like uh GE, ABB, all of those big companies and we went through that process and it it was a whole year-long process. uh went through multiple rounds of you know assessments and validations by the utility and at the end we came on

top and were selected as their vendor for AI enabled infrastructure inspections. So that was a great moment you know landing your first customer like that you before that we only had ran like PC's and pilots smaller pilots but landing New York Power Authority in 2022 through that RFP was was a great achievement and at that time we knew

that this is this is going somewhere. What? So in that in that year though, you know, what what prepared what in in retired you for success when that opportunity came for the RFP? I think one of the key things was since we spent so much time on building a really awesome product and a product and

technology that works. So one factor was we we invested heavily in building a really positive and a really accurate uh product that utilities liked and at the end of the day the the reason why we won the RFP was because of our product. So they the utility New York Power Authority did at the end they had like

10 vendors left and they did an assessment uh which we call an AI bake off which was they gave us a blind data set and gave it gave everyone 30 minutes to process analyze the data set. So that 30 minutes because they don't want anyone to manually kind of go through it imagery. So what we did was we returned

the data set with results back in seven minutes and the rest of the people took more than or rest of vendors took more than 30 minutes but also our accuracy metrics were way higher than anyone. So we were at 85% baseline accuracies of our AI models at that time and the closest was 32%. So that was a massive

massive accuracy jump that we had that kind of helped the utility to you know choose us. So one aspect was the was the product really worked cuz we spent so much time. The second was the product worked because we took a bet early on on an industry which was still very new. So

AI for asset uh or grid in infrastructure inspections was still very new and was not proven as a as a market in when we launched. So we took a bet on that and we didn't know. So it took the industry 5 years to get to where it had to go. So we had to wait 5

years for the industry to catch up to the technology and this is even before the boom of the large language models. So that that also kind of worked out. So what so what is it then to the capacity you can speak on it what is it that differentiates your you know for for

someone that doesn't have super technical uh depth in AI or ML and how they're constructed how they can differentiate uh on a on a deep technical level what differentiates your approach from uh the other um you know industryleading competitors that you had at the time that were taking longer to process the data and had lower accuracy

was it just a longer a longer exposure to your product to utility data or was it something else? Yeah. So, you know, obviously cannot talk a lot of like trade secrets and what's going in our IP and there's a lot of, you know, processes and the science behind it, but on a high level, I would

say we focused on data. Data is the key for any kind of AI architecture. And the common saying is garbage in garbage out. So you want to make sure yeah you want to make sure that you don't throw garbage into the systems. So you know what worked for us really is early on because of that three four year

period where we were building the product that was the R&D period for us cuz we were we were getting access to the right data sets and that could go in and the quality data set quality data sets that can go into really mature systems. So the first piece was we figured out the data piece. If there was

another Silicon Valley startup that comes in today and tries to build this uh they won't be able to because there's no publicly available data uh for this critical infrastructure that utilities have. So they cannot go in. It takes time to get access. But at the same time we were at the right place at the right

time where utilities were still new figuring out how to manage the data and we had a promise that we can help them manage the data. So that really helped early on to build our products, do our research and development, align with the right partners. And once we figured out the data problem, we went into the AI

problem, which is, you know, building the right architectures. We don't use any off-the-shelf models cuz they don't work in this space cuz there's just so many different variables that we have to take care of when we are building these systems. So we build our own proprietary models and and train them over the years

and now they have become so comprehensive and advanced uh that it's really helping utilities at scale. So it was yeah it was it was a combination of you know the product being data and AI but also the relationships we build early on. So shi shifting uh a little bit uh into I mean another critical

piece of building the company. How what what have you learned you know what can we take away from those early years like you said the first five years for the for the uh industry to come around and now you guys are are finding a lot of success. Um, but what what can we take

away from your co-founder dynamics or if you guys had early hires after your seed round uh around how you built the culture that that enabled and catalyzed the the patience, the persistence uh in order to build your product.

Yeah. No, certainly. And uh you know entrepreneurial journey is all about dynamics within the team and you have to facilitate a positive team culture. But you know early on so my co-founder she has a business background. She comes from economics, finance and I have a technology background. So it was obviously a really natural fit. Uh which

was you know business and technology working together. Uh so that was the first kind of team dynamics that really worked out. Um sometimes the dynamics they don't work out cuz there's clashes in in the skill sets but in this case there's a natural fit. Uh then we started building the team. Obviously we

hired after the seed round we hired some critical engineers to keep building the product make the product better cuz you know with utilities obviously relationships helps but in our case the product was dictating it. Utilities want to see good solutions they had been promised a lot uh you know we had competitors at that time that you know

went out of business because their product didn't work. There was a lot of overpromising and underdelivering. We believed in delivering uh quality products to the utilities. So that really helped. We brought in right engineers at the right time. Then we started bringing in after we raised uh further rounds of financing. Hey, now we

have the product nailed down. Let's let's bring in sales operations into the mix. So then we started hiring sales resources, business resource resources because once we have done it a few times, sold uh led the founder sale with utilities a few times now we can replicate that with the sales team. So that was our journey and in terms of the

team culture we really really value um ideas uh we we keep obviously keep an open door policy of full fully remote team uh and team is spread out across the country and it's been working out for the last 5 years but but the the key aspect is building team dynamics um and then team culture. So yeah, obviously

keep an open door policy. We are fully remote teams, teams spread out across the country. We always value ideas. So we we think that you know transparency and ideas are are key to building uh really great team cultures as well.

So are are there any um and I do want to get to you know post uh industry coming around post commercialization things like that you know the the acceleration of the company your experience doing that because I also think that's valuable. Are there any um I don't know let's say whether it's in the the hiring

uh or the the team building or whether it's within product decisions were there any major mistakes that you made that you look back on that that really shaped the way that you navigated the company after that? Yeah, I mean um growing a venture from you know 0 to one is always very very

tricky and you know mistakes happen and you learn from them. Um you know there's there's a saying that 90% of startups fail. So you know there's always risks involved. So you know over the journey we we definitely had made mistakes cuz we are first-time founders and we're learning things on the go. We have a

good good group of advisers. So that's another thing that we did was we we collected a good group of advisers to advise us from like day one uh to tell us from their journey of starting companies and scaling and uh acquiring companies and from the investment portfolios as well to tell us you know

in certain critical scenarios how to navigate those. But you know as hiring is always tricky um you know even with larger companies and obviously with startups early hires mean a lot. Um, you know, we've all obviously went through the hiring phases of we we've had some bad hires, but with the critical thing

is, you know, uh, it's it's all about the fit of the of the hire of a candidate. So, we were able to kind of assess that early on and kind of, you know, course correct accordingly. And now we have an awesome team of people that are really committed mission and the P and they're passionate about the

energy and AI sector. So, but we've had our we've had our uh you know bad fits in the company as well that uh that didn't fit well with the company and we were able to kind of navigate through that and learn.

Well, yeah, just I mean just learning from something like that, you know, from let's just take just very simple example, you know, the the first hire that you brought on and then let go. What was the the main thing that you changed about uh your hiring procedure after that?

Yeah, it's it's so interesting. Um obviously our hiring procedure in terms of you know how we run the interview rounds evolved over time. Y uh you know for technical hires you know coding challenges interview rounds uh evolved even for you know nontechnical candidates like sales and business that evolved over time. So we can see

literally 5 years ago how we interviewing and now there's much more process and framework into it. But something that me and my co-founder discuss all the time is never underestimate the power of the gut feeling which is the intuition. Uh and we learned that the hard way cuz sometimes our intuition would be off

about someone and then it tended to be that yes the intuition was right about it. So over the years we have started trusting our gut feeling and intuition uh about a candidate and their fit with the company. So that's one lesson that we learned over the years is don't you know never underestimate it under

underestimate the power of intuition and gut feeling. Has has there been any any um has there been any misstep I guess that you would look back and say as far as you know with with working with utilities and things like this? I mean these are not there's not like a ton of them out there. It's not like you're,

you know, working with customers, coming on, dropping off. Like you mentioned, there's a huge process to get involved and then you stay involved um with that engagement with the utility. But were there any engagements that you had that were uh either not a good fit because of where Buzz was at in your development

cycle or not a good fit for Buzz as the the type of customer that you want to service? Yeah, definitely. I think we've had over the journey of of the company we've had a few customers that were um we had that they they wanted a lot and there was a drain but there was also not a drive to

kind of excel their programs internal programs and we've had customers that were interested in the product and technology and they wanted to um you know use us but they lost budget due to some you know uh you know external factor uh you know maybe they had a major disaster that happened that made

them lose budget and then that's why they couldn't use us. But from from a customer standpoint, I think we we learned a lot to when we are doing the pitching to customers or selling to customers is to understand a lot of different features is first of all, you know, if they have the right budget, are

we talking to the right people? You know, the common sales enterprise sales principles. But that came we learned a lot uh over the years uh cuz obviously we had never did enterprise selling uh but over the years we learned how to do enterprise selling and to the to the level that we can sell to enterprise

enterprise customers multi-million dollar contracts right so but that involved a lot of pitfalls. Uh so the answer is yes. We had customers that had no intention of buying us long term but they want to keep on like wasting time and we learn to assess that uh assess that over the period of time as well.

Yeah those are those are those are tough experiences to go through you know cuz like we're talking about like the the the ups and the downs of the journey. Um, you know, I think one of one of the craziest things, one of the hardest things I you know, I think in the journey is to acquire your first

customer, acquire your first couple customers, especially when you're doing something novel like what Buzz is doing. And uh to have one or multiple of those customers end up being really difficult engagements, whether it's some, you know, uh the the the person that you're working with or because it's just genuinely not a good fit. and you don't

know that yet as a founder because you're still learning like you said you know those are really tough experiences to go through so I'm glad to see that you guys you know made it to the other side so from um as a founder in the space and your uh your experience with AI ML things like

this I think you have a obviously really good um pulse on the technology on the industry where it's going and it's also interesting that that you have to be in uh energy as well So if you could just give a how has the industry uh since that that that threshold of when they

were ready uh to engage with buzz how has the industry changed um and and why you know outside of this whole data center in the mainstream uh yeah so there's been a massive change in the industry over the years so uh you know 20 25 years ago utilities were still inspecting their power lines and

power infrastructure with with helicopters and they were manually kind of going through reports and looking at this infrastructure. But what has changed over the years is you know the weather has gotten extreme. Uh there's more stress and strain on the power grid uh due to electrification.

You know renewables coming online. There's electric vehicles coming online and industries becoming electrified in general. And then recently now we're seeing the data center uh load being massive driver for utilities to get more capacity on the grid. So what that requires is you know the grid is think of it as a highway of of electrons and

now it's starting to be two-way and it's starting to get congested. What that means is that the wires degrade much faster. The equipment electrical equipment degrades much faster and then you have the the consequences of like weather conditions that are you know ve vegetation on the line that can be a fire risk or like higher drought or

higher heat uh kind of climatic conditions that could be a fire risk. And we've seen wildfires spark due to fail grid infrastructure. So what it needs is utilities need to inspect the infrastructure more frequently. And in order to do that they need to employ more innovative tools and technologies.

So here comes the drones. In the last 10 years with drones they can inspect much more infrastructure much more frequent in infrastructure and then easily and save a lot of cost make the operations efficient. But that also means that the data that they collect is is becoming 10x. So, and just to give you context, a

utility like PG&E collects like 50 million images annually. Uh, so that's a lot of data for utilities to manage. That's where AI automation tools can come in and that's where we come in. We take all that data in, manage that for utilities and tell using our computer vision AI, we tell them what kind of

problems are happening on the infrastructure, whether the lines are overheated, uh whether insulators are damaged, whether there's vegetation on the lines, those kind of things. So what is in in this space of uh let's say utility how would you describe it like utility uh damage control like what what's what's the label for this this

space it's it's called so there's few labels so this is more on grid inspections but the the label is asset management and we trying to provide the next generation of asset intelligence for asset management and grid optimization for utilities.

Okay. So, so, so in the space, are there any um uh are there are there any arguments against the use of drones or helicopters or images or is this truly the uh the industry standard for like are there any other solutions that utilities explore?

I think utilities are at a place where they need all the tools in the tool shed to apply to this because these are really complicated problems that traditional methods that were employed 20 years ago are not working out and that's why we're seeing you know wildfires happen there's storms that are uh creating a lot of problems massive

power outages happening so they need all the sensors so utilities are deploying multiple IoT sensors on their infrastructure to collect more data um drones and drones are just a mobile sensor right There's a flying sensor. So, they're deploying more drones on the infrastructure. And recently, we're working with utilities that are deploying drones in a dock, housed in a

dock at their substations that can autonomously fly and inspect uh infrastructure. And that's kind of the future of inspections. But that also comes with a lot more data generated from these mobile sensors. So, I think the future of inspections and monitoring is for things to kind of work out for utilities. They have to deploy all these

sensors even more frequently and more volume but then they still need to solve the data problem and that's where companies companies like ours come in to help them analyze this data. Are there are there any are there any common misconceptions that utilities or prospects have uh of your solution that make them not interested in engaging?

Um I I think it's more about education because utilities are inherently not AI companies and they are still starting to learn how AI can be positively applied and and there's a lot of innovation happening. So there's and we have done that. We've done uh we we've become a thought leader in this space cuz we were

so early in this space right launched 5 years before the market caught up. So we started putting out how to guides on how to evaluate AI in this space, how to deploy AI, how to make sure your data is secure and cyber secure because this is critical infrastructure data that cannot be leaked, right? So we also kind of guide

them on that journey. So we we're trying to be their their one-stop shop for not only product but also consulting in some ways on how you can deploy these programs at scale. So then so then quickly uh if we could just go over since um since that threshold of commercialization what what is your and you know alongside

the the insane uh growth of AI adoption uh in the world and subsequently the need to feed uh those data centers and processors and software and hardware and everything. Um what what has the experience been in the last uh four years? Have you guys been rapidly uh adopting new customers?

Has that adoption timeline by utilities shortened at all? Um you know, you guys have grown your team, you know, how has that been from a cultural standpoint? You know, any any of these directions? Yeah. No, adoption has increased a lot for uh for the utilities that we've worked with. One thing about utilities

is no one wants to be the first and second customers for a new technology, but everyone wants to be the fourth and the fifth. Uh so great thing for us was we were able to land and successfully showcase our technology to the you know first couple and now we have utilities that we have to schedule out their

projects uh because we just have so much demand that we can take in cuz every utility is facing some kind of problem whether it's you know power outages or wildfires or storms or you know snowstorms those kind of things and and the only options that they have is they have to innovate they have to apply

these cutting edge technologies AI boom has happened in the last you know four or five years and we've also kept up with that we have adapted our technologies accordingly and now what I'd say is you know data centers so in order to feed AI you need power that's the missing piece and utilities are

struggling to provide power to data centers but AI technologies like ours is also helping them unlock more capacity and more power that they can supply to data centers so it's like a closed loop AI helping more you know evolution of AI Yeah. So, so that's where a lot of adoption is happening as well and that's

where I think a lot of our solutions can fit into data center powering as well. Uh utilities are at the you know cross-sections of they need to use AI for add scale for various kind of use cases otherwise they'll be left behind.

So my two two of my favorite uh questions ask my first one is from your perspective what is the biggest hurdle that you're facing at the moment? How is it also an opportunity? Yeah. So the biggest hurdle is obviously um you know deployment of the AI at scale for utilities but also the selling

to utilities takes a lot of time. So from business perspective utilities are as I said they're slowmoving behemoths. Um they're they're massive organizations. You have to go through multiple different processes and frameworks with utilities in order to get adopted right. You have to do cyber security review. You have to work with their procurement, their sourcing. You

have to prove business value, business case, do demos, run validations, IT governance, all of those things. And that takes a lot of time working with utilities. Uh especially the major ones we're talking about, which are investor own utilities. Uh and there's 250 of them. And then there's other smaller utilities that are around like 4,000 of

them. So, so working with them, the sales cycle expands. So on an average selling SAS to utilities is 12 to 18 months of sales cycle. So that's a hurdle. We have used innovative techniques like channel partners and all those to shrink those sales cycles but it's still as compared to other industries still a long sales cycle.

However, that there's an opportunity also. So once you get into a utility and once you sell to utility, the churn risk with that utility is very low cuz they've gone through so many processes with you. Unless you mess up really badly, they would stick with you and you build that reputation and and kind of uh

you know a relationship with that utility. So there's opportunity over there. Um also utilities do sign multi-year contracts with you. So that's another great thing for for businesses. So in terms of like the business side of things, the hurdle and the opportunity work side by side with utility selling motions. Um and if I were to think

generally I think just education like there's as I said there's still a lot of misconceptions about AI and that's a hurdle in this industry but that also gives opportunity opportunity for a player like ours to come in and become that thought leader and that's what we're trying to do because we have had

that experience with the AI programs and cycles and how to scale those how to effectively deploy those and that gives us kind of a an advantage of having a first mover advantage early on to be in this sector and kind of scale that with our utility customers.

What what are those uh misconceptions that you've seen? Yeah. So I think over the years there's as I mentioned before there's been a lot of overpromising and underd delivering and there that casted a lot of doubts in utilities minds that does this even work uh right but there's a certain way of making it work and there's certain

processes to follow with any other tool people think AI is everything but it's at the end of the day it's just a tool right so the way to use a tool is to be trained on effectively on how to use the tool but also how to interpret the outputs of the tool or the results from

the AI right so there's education required for that you know when the when the internet came on as as a tool there was you know people had to get to use it they had to get to train it emails and all those things same as with AI softwares and and the value of AI that

needs to be understood at the at the low level for for utilities um the the higher push from the executives is heavily on the AI but what our goal is to train engineers, electrical engineers, power engineers, field technicians to train them to understand that this is a tool. This is not causing a loss of job for you. This is another

tool in your tool shed that you'll use. You know, you'll use wrench and all those hardware tools, but this is a digital tool that you can use to make your job much more efficient and much more safer. Well done. That was that was great.

That's that's really useful. Um and uh my last one for you is uh with all this work to be done uh these long sales cycles, these misconceptions about um AI, you know, navigating uh leading an increasingly big uh company and all this stuff. What inspires you?

Um obviously Buzz as as my company uh really inspires me. The team that we have built at Buzz inspires me on an everyday basis that makes me wake up in the morning, go to work, do do my job to the most optimal level. But also, I think the future um thinking about the

future of technology really inspires me cuz you know we have so much great technology at hand and and we living in an age of solutions, right? We're living in an age of problem solving. you can literally use a large language model and find a solution of the problem really quickly. So, how can we leverage that to

kind of enhance our day-to-day operations, make them much more efficient is really critical as well. So, I'm excited and inspired by where AI is going. There's a lot of doubts from people that, you know, this is going to be used in bad ways. And obviously with any other technology, yes, there will be

areas that it will be used in a in a really negative manner. But there's a lot there could be a lot more positives about this new technology that that is being developed that can actually really solve real world problems that are not able to be solved with conventional methods that we have tried before. And

these are really complicated problems. uh you know some of them might be man uh human created but at you know what we need right now is technology to be infused at that right level to solve those critical problems um and also I would also say regulatory commissions to catch up with that uh technology is

always ahead of regulation but I think it's time for policies and regulation to catch up with technology and make sure they're the right standards are applied so then technology cannot be used in a negative manner so that inspires me out the future there's a lot of opportunity and obviously hopefully Buzz can help with

that as well. That's right. May solutions making the world a better place, all this stuff. It's good momentum. Vic, I appreciate the energy uh during this conversation. I'm excited for our next one when we get an update and get to dive uh you know this very we we we hovered on the surface I think a lot.

Um uh but but it was still really fun. And if anyone else was uh inspired to follow along with your journey, what's the best way to do that? Yeah, please follow me on LinkedIn. Uh I'm uh Vikat Shri on LinkedIn and happy to share more anecdotes about my journey uh some pitfalls, common mistakes, risks

we have taken. I can dive deep into that and what's the future of AI. Would love to have a chat about that also. So follow me on LinkedIn and connect with me on LinkedIn. Do it. Do it. connect with him. It's super fun.

Awesome. Well, thank you so much for your time and uh yeah, looking forward to the next one. Yeah, looking forward to it. Thank you, Blake. Appreciate your time as well.