The CEO of Everguard.ai, Sandeep Pandya, joins Ryan Chacon to debate how AI advances IoT use instances and the way it can save lives. Sandeep begins with a sharing the backstory of the corporate after which dives right into a high-level overview of the present state of AI and the main use instances of adoption. Sandeep then focuses on the significance of AI and its potential to guard staff’ security. Ryan and Sandeep wrap up this dialogue with a dialog round challenges to adoption and recommendation for corporations seeking to combine AI into their enterprise.
About Sandeep
Sandeep Pandya is CEO of Everguard.ai main the way in which within the incorporation of wearables with pc imaginative and prescient, machine studying, and AI to make the world’s industrial environments safer and extra sustainable. Shaped in 2019, Everguard.ai stemmed from Pandya’s perception that the world is experiencing a technological renaissance and that the nexus of AI, digital connectivity, and knowledge sharing will essentially reshape and enhance how individuals and their communities maintain one another. Pandya is a know-how veteran whose expertise spans product introductions and improvements in semiconductors, related gadgets, wi-fi infrastructure, cloud providers, and novel AI/CV purposes. His management has helped organizations of all sizes set up high-growth markets globally.
Thinking about connecting with Sandeep? Attain out on Linkedin!
About Everguard.ai
Everguard’s mission is to guard corporations’ most vital belongings — their individuals — with the primary proactive know-how answer devoted to industrial sustainability. They’re revolutionizing how heavy trade approaches employee security, well being, and welfare utilizing AI know-how to assist corporations of their missions to satisfy environmental, social, and governance (ESG) initiatives. Everguard.ai are the technologists behind Employee-Centric AI™, the world’s first synthetic intelligence platform to create a aware setting powered by sensor fusion that senses misery and hazard lengthy earlier than a human can. Employee-Centric AI™ ties collectively varied industrial sensor applied sciences utilizing sensor fusion, edge computing, and AI algorithms to allow them to carry out in methods not attainable independently. Everguard’s enterprise answer, Sentri360®, makes use of Employee-Centric AI™ know-how to create a Sentri Zone™ with staff on the middle, constantly defending them in probably the most advanced industrial environments.
Key Questions and Matters from this Episode:
(1:51) Introduction to Sandeep and Everguard.ai
(5:49) Present state of AI
(9:24) Use instances main AI adoption
(11:23) AI for saving lives
(14:43) Challenges to adoption
(18:21) Recommendation for corporations seeking to undertake
Transcript:
– [Sandeep] And principally present telemetry into software program that now AI software program, whether or not it’s working on the edge in that little digicam that’s understanding that the gentleman just isn’t sporting his visor subsequent to the glass furnace or it’s aggregating hundreds of thousands and hundreds of thousands of information factors, you get productiveness use instances ’trigger now you possibly can see, you understand, operational bottlenecks earlier than they occur.
– [Ryan] Hey everybody and welcome to a different episode of the IoT for All Podcast from IoT for All of the primary publication and useful resource for the Web of Issues. I’m your host, Ryan Chacon. Should you’re watching this on YouTube, we would really respect it in case you give this video a thumbs up and subscribe to our channel, you probably have not completed so already. Should you’re listening to us on a podcast listing someplace like Apple Podcast, please be at liberty to subscribe to get the most recent episodes as quickly as they’re out. All proper, on at this time’s episode we have now Sandeep Pandya, the Chief Government Officer of Everguard.ai. They’re an organization who has a mission at defending firm’s most vital belongings, that are their individuals, with a primary proactive know-how answer devoted to industrial sustainability. Very attention-grabbing firm. Positively advocate you verify them out. On at this time’s episode, we speak lots about AI, we speak in regards to the present state of AI, the function AI has within the IoT house, how AI is even enjoying a task in saving lives with various kinds of purposes and use instances, massive challenges that corporations face once they’re attempting to include and undertake AI into their answer and in addition common recommendation for these seeking to carry AI into their IoT options. So improbable episode total, I believe you’ll get plenty of worth out of it however earlier than we get into it, any of you on the market seeking to enter the quick rising and worthwhile IoT market however don’t know the place to begin? Try our sponsor Leverege. Leverege’s IoT options growth platform, offers all the pieces it’s worthwhile to create turnkey IoT merchandise you can white label and resell beneath your individual model. To study extra, go to IoTchangeseverything.com. That’s IoTchangeseverything.com. And with out additional ado, please take pleasure in this episode of the IoT for All Podcast. Proper Sandeep, welcome to the IoT for All Podcast. Thanks for being right here this week.
– [Sandeep] Yeah, thanks for having me.
– [Ryan] Completely. All proper, let’s kick this off by having you give a fast introduction about your self to our viewers.
– [Sandeep] Thanks Ryan. I’m Sandeep Pandya, I’m the CEO of Everguard.ai. Everguard.ai is a enterprise backed startup positioned in Southern California, targeted on utilizing deep applied sciences like synthetic intelligence and industrial IoT to make industrial staff safer all over the world.
– [Ryan] Incredible. And what I’d love to listen to, in case you wouldn’t thoughts, is sharing slightly bit about kinda how this firm even got here into existence. So, you understand, I do know there’s a singular story about form of how y’all began, however what was the chance you noticed, what have been you form of seeing not being served available in the market, what introduced this firm into existence and form of obtained you to the place you at the moment are.
– [Sandeep] Yeah, thanks for that. So, you understand, I’m a know-how veteran. I’ve been doing know-how product growth for over three many years, principally at giant blue-chip corporations however again in 2015, I broke away on my first enterprise backed startup, it was targeted on employee security utilizing deep applied sciences however within the business fleet house and as president over there, I assume I used to be recruited by one of many traders of Everguard, which is Boston Consulting Group after which their accomplice SeAH, which is a big Asian metal conglomerate to principally take an analogous idea of utilizing deep applied sciences for employee security however within the industrial setting, significantly metal mills which SeAH has they usually have been in search of an skilled government who might take seed capital and construct the group, the tech and launch the product within the firm in order that not solely will SeAH staff be safer, however you understand, working staff in all industries may be safer. And for me, issues are going nice at my prior firm, they’re on their technique to turning into a unicorn. I really feel nice about that however the alternative that basically excited me, Ryan was within the pitch deck from Boston Consulting Group to me, which confirmed that OSHA, proper? The physique that regulates employee security has discovered by their analysis that the highest 10 incidents that happen on the job in an industrial setting typically are the identical throughout all industries and to me that was an aha second, which stated that, and this was the pitch from BCG to me, which is, “Hey, come and be the CEO of this firm and in case you may also help us construct a security platform, an AI security platform for metal, we will go to grease and fuel, we will go to development, we will go to mining” and that’s actually what we’re attempting to do right here, is simply attain extra individuals and so to be part of that to outline employee security in a brand new manner, so broadly, so shortly was simply one thing I couldn’t say no to, so…
– [Ryan] Yeah, I believe the, you understand, that pitch they needed to you, clearly it labored, however on the similar time, it’s very related to the businesses that we’re seeing through success within the IoT house proper now once they’re in a position to construct for a specific drawback but it surely carries plenty of weight throughout industries and it’s one thing that could be a frequent drawback, not simply in a single siloed vertical however throughout many alternative industries, and it lets you have an actual alternative for scale and I believe these corporations who’ve these varieties of options, have these varieties of merchandise, are seeing plenty of success within the house that we, you understand, we give attention to.
– [Sandeep] I fully agree. I believe the superpower that IoT has whether or not it’s industrial or in any other case, is absolutely it’s horizontal scaling energy, to your level. Which is, it’s plenty of, you understand, primarily nicely understood know-how but it surely simply must penetrate completely different areas and once you form of dial it in a single or two locations, one or two verticals, or market segments, in a short time, in case you’re sensible, you possibly can take it sideways.
– [Ryan] Completely. So let me ask you at form of a excessive stage right here, I do need to dive into extra of how AI is enjoying a task in saving lives and so forth, however I needed to begin slightly additional out and talk about the present state of AI. To lots of people once they hear AI, they hear it as a buzzword, they actually don’t perceive precisely what it does and what it’s doing so as to add worth to options which can be available in the market. However in case you might begin off by simply telling us slightly bit about the way you view the present state of AI generally after which additionally the function it truly is enjoying otherwise you see it enjoying within the IoT house.
– [Sandeep] Yeah, nice query. And you understand, by itself it may very well be a really lengthy dialog however within the easiest phrases, you understand, I’ve been doing AI on and off because the late 80s and early 90s, you understand, after I was in grad faculty and that’s when AI wasn’t practically as cool as it’s at this time. And it was additionally much less mature, you understand, and the utility of it was lots much less as a result of it was guidelines primarily based programming, you understand, if A and B then C. And compute energy was lots decrease and so you actually couldn’t remedy something however what they name an academia toy issues. Now what has occurred is, you understand, during the last three many years and so, plenty of different enabling applied sciences like within the smartphone world, have miniaturized processors, cameras, location applied sciences like GPS and principally created a way more attention-grabbing and reasonably priced set of platforms upon which these candidly compute intensive algorithms can run and in order that’s one factor which is, it’s gotten cheaper to do actually, actually refined programming. However the different form of, I’d say inflection level within the AI world was the arrival of deep studying which is a neural community sort of programming for those who could learn about it, but when it sounds too fancy, it’s actually only a completely different manner of attempting to put in writing algorithms that work higher for issues like picture evaluation or audio streams or any form of sign that’s not so simple as A + B = C, proper? Our conditional form of stuff and so actually what that did is it created a renaissance in pc imaginative and prescient as an illustration otherwise you may need a sensible speaker at dwelling, so pure language processing, it simply, whether or not it’s Alexa or Google, it simply appears a lot extra intuitive now and that’s often because a couple of decade in the past, deep studying kicked in and made understanding human language, understanding human exercise, which means is the particular person smoking? Are they placing one thing of their jacket in a retail setting once they shouldn’t, you understand, simply these varieties of human behaviors far more simply understood. Now, I’m not saying that the programs are clever, they’re simply actually higher at info evaluation when the data is that this unstructured form of picture or audio stream and so anyhow, that’s slightly bit nerdy, however the enabling power is as a result of it’s turn out to be cheaper to deploy within the industrial IoT world as a result of the processors are cheaper as an illustration or mobile connectivity is extra pervasive. And the algorithms themselves are completely different than 30 years in the past, that good storm is now made AI far more approachable and also you may need a hoop doorbell that may inform it’s a porch pirate versus a automobile passing by, proper, or the sensible speaker can perceive what you’re saying and order the appropriate factor, et cetera.
– [Ryan] Completely. Are there, because it pertains to what an AI is ready to, or is enabling inside IoT, are you seeing any use instances form of main the way in which in AI adoption or one other, I assume one other query form of associated to that is how AI is advancing sure use instances or any specific use instances that you just’ve form of seen actually form of bubble to the highest lately?
– [Sandeep] Yeah, I’d say even outdoors of the security world, one of many powers of IoT I believe is extending connectivity to locations the place earlier than you simply couldn’t, so whether or not you’re on the store ground of a fabrication facility or within a bar mill or on a development website, you understand, these are actually difficult environments wherein to function, proper? They only are. And the ability of the IoT neighborhood is to determine how you can create gadgets which can be related, that may do work, like perhaps sense issues, sense the state of a machine, sense the air high quality, you understand, et cetera and even human conduct by pc imaginative and prescient and cameras. And principally present telemetry into software program that now AI software program, whether or not it’s working on the edge in that little digicam that’s understanding that the gentleman just isn’t sporting his visor subsequent to the glass furnace or it’s aggregating hundreds of thousands and hundreds of thousands of information factors you get productiveness use instances ’trigger now you possibly can see operational bottlenecks earlier than they occur. Or you possibly can see that the employee is trying the fallacious manner when the crane is approaching. Proper. Now you can create a world of what we name not simply predictive analytics however what we name prescriptive, which implies I really not simply am predicting one thing unhealthy will occur or unintended, however I can prescribe the answer in actual time, in tons of of milliseconds and IoT is that spine that allows that, the compute plus the IoT really makes all of that occur.
– [Ryan] Incredible. So I needed to form of pivot in the direction of the main target space that you just all have as a result of I believe it’s a really attention-grabbing house that we haven’t lined all that a lot and it’s round using these applied sciences that we’re speaking about to avoid wasting lives and significantly because it pertains to employee, that staff who work in these difficult environments, so development, industrial manufacturing, you identify it. Are you able to simply speak slightly bit about how AI is absolutely getting used to assist defend staff in these environments and simply form of in a common sense?
– [Sandeep] Yeah, no, I really like that query and we at Everguard have created a framework of serious about these advanced environments ’trigger if anybody’s walked the ground of a metal mill or a development website, it’s as busy because the busiest airport, proper? I grew up in Chicago, we all the time took delight that O’Hare Airport’s the busiest airport on the earth. I’m positive some individuals will argue with that, however the backside line is these are extremely advanced environments. Individuals transferring autos, machines like cranes, the air high quality itself, after which the operation which isn’t allowed to cease if it may be helped. So we have now a framework that breaks it up into 4 buckets and AI exists in any respect of them. So there’s the employee stage, boots on the bottom, proper? There’s the workspace, that’s the man or lady to machine interplay. So automobile to employee, crane to employee, machine to employee, you understand, it may very well be machine guarding, excessive pace machines. Then there may be the work setting. So OSHA as an illustration, because it pertains to security, perhaps individuals know this or not, but it surely’s not nearly what the employee is sporting or how they work together with the machine, however the air they breathe, the sunshine air pollution, the noise air pollution, all of which have very strict necessities to maintain the employees protected and so it’s worthwhile to monitor that with the appropriate varieties of sensors after which finally, you understand, when you will have the employee, their core physiology, PPE, the workspace, their interplay with machines and the work setting like environmental sensors air high quality as an illustration, or noise or mild. All of that telemetry aggregates into what we name the company governance layer. In order the ESG motion, as you understand, Ryan, is all about sustainability and firms doing the appropriate factor to maintain staff and their communities protected and healthful. How do you do it? The place’s the info coming from, proper? It’s utilizing paper and pencil whereas individuals strolling round manually. Nicely, within the twenty first century, that’s loopy. However IoT plus AI, what it permits is it actual time telemetry to movement as much as the governance stage, so the C-Suite, the board of administrators, shareholders, they now know that once you have an effect report from that massive operation that there’s credence, there’s credibility to it. And so we’re an enabling power, I believe IoT and AI can present that actual time telemetry after which once you use it the way in which we’re utilizing it, which is prevention of incidents, whether or not they’re environmental or employee or damage associated, then you will have the power stepping into two instructions.
– [Ryan] Completely. Yeah, that’s improbable to form of break down for us and really highly effective form of as you possibly can clarify right here, how these applied sciences are enabling issues that weren’t attainable earlier than? In these deployments that you just’ve been concerned in and the, you’ve seen corporations undertake these applied sciences and implement them, what are a number of the greatest challenges that you just’ve seen when it comes to the adoption aspect of issues? Like once you work with an organization to deploy an answer what challenges are corporations oftentimes operating into and the way ought to individuals be serious about it who could also be listening to this in order that they’ll form of get forward of it once they’re planning for the implementation of such a know-how?
– [Sandeep] Yeah, I’d say, two buckets. I prefer to suppose when it comes to frameworks and so there’s one body which is the maturing of the know-how in your specific house and so what I imply by that’s, and folk which can be from an AI world or a tech world with some familiarity round AI, AI principally trains itself by further knowledge to get an increasing number of exact, proper? I can use the phrase studying in case you wanna give it a, in case you wanna personify the conduct however principally the extra knowledge you feed an AI algorithm, the extra exact it will get. Facial recognition or you understand, no matter, what have you ever, proper? So after we drop some new use instances in, you understand, to watch PPE or the way in which individuals are interacting of their setting, despite the fact that we would have completed that use case 5 different occasions in 5 different amenities, AI is slightly bit delicate. That means if the lighting adjustments, if the the background litter is just a bit bit completely different, there is usually a little little bit of a startup value to retrain the fashions, what we name fashions, the AI algorithms, to guarantee that the KPIs, that the important thing efficiency indicators are within the excessive 90%. So simply coming in, it may be 70% or 80% however with slightly little bit of retraining to make it bespoke to your setting, it’s again within the 90%, but it surely wants slightly little bit of time. Then again, there’s another applied sciences like automobile to employee anti-collision programs which can be radio sign primarily based, they simply work out of the field. It’s simply radio know-how, proper? In order that form of nuance, little bit understanding of how you can deploy and the endurance is useful. And we’re speaking about weeks and months, not months and years, proper? The opposite framework is the change administration piece, the organizational piece, which is, a few of your viewers may say, “Yeah, that’s all nice however you understand, we’re in a union store, the fellows are by no means gonna settle for cameras in there” and I’ve completed this now for eight years after being a deep tech man in massive corporations. For the final eight years I’ve been doing employee security by AI and first in business fleets the place drivers didn’t need cameras of their faces and you understand, now on store flooring and metal mills and manufacturing amenities. What we have now discovered since 2015, is that individuals, consider it or not, are extra snug with AI, it’s much less mysterious, individuals have a sensible speaker, individuals use their smartphone, they’ve ring-
– [Ryan] They don’t suppose it’s gonna take over the world or any, or like-
– [Sandeep] Precisely. I prefer to joke, it’s not Arnold, it’s Alexa, calm down. And for that cause there’s slightly bit of hysteria however consider it or not, we’ve talked to a number of the largest unions on completely different continents they usually get it they usually perceive that this know-how whereas it seems creepy, it’s not meant to be, it’s not smarter than it seems to be, it’s actually simply educated to do particular issues, like is that particular person sporting a hat? It doesn’t know that they’re smoking a cigarette in case you haven’t educated it to try this, proper? It’s not watching you in that manner. So, lengthy story quick, I’d say that the change administration piece looks as if a barrier but it surely’s not likely in case you simply prepare the group, speak to ’em and the 2 issues collectively assist us roll out.
– [Ryan] Completely. And I assume to kinda add onto that as one in every of my final questions earlier than we wrap up right here, what recommendation do you will have for corporations seeking to carry AI know-how into probably even current IoT options or to assist remedy different issues that till now actually weren’t attainable for them? What sort of recommendation do you will have for them to form of how you can strategy it?
– [Sandeep] I’d say simply very merely the know-how is prepared. We perceive that change takes time, so get began now, you don’t have to spend so much, there’s corporations which can be keen to work and begin small. One factor I realized in enterprise faculty is an previous phrase, “Trial results in adoption”. Do a trial or two, get everybody snug after which actually begin making your environments higher and smarter with this tech.
– [Ryan] Completely. Yeah, I believe it’s a quite common piece of recommendation, particularly with the conversations I’ve had within the IoT house when you understand, begin with a trial, perceive the way it’s gonna be integrated into your enterprise, into your legacy programs, your current infrastructure, begin seeing ROI as a way to then justify the means to deploy it at scale after which that’s form of clearly the place we’re all attempting to get to. So improbable recommendation. Completely. So for viewers who’s listening to this and desires to comply with up, study extra, dive right into a bit extra about what you all do, the function you play within the house, perhaps comply with up with any questions, what’s the easiest way they’ll do this?
– [Sandeep] Yeah, we’d like to area any questions, even when it’s simply to your personal edification, you simply wanna study extra, we’re blissful to speak to you. [email protected] and Everguard is rather like it sounds, E-V-E-R-G-U-A-R-D.ai [email protected] and we’ll leap throughout it and get again to you.
– [Ryan] Superior. Nicely thanks a lot to your time. Incredible dialog, our viewers I believe is gonna get a ton of worth outta this. We’ll ensure we hyperlink up all the required info within the descriptions and the content material we push out round this when it goes dwell, however till then, thanks once more and hopefully we will have you ever again in some unspecified time in the future to speak about different matters associated to what you will have happening.
– [Sandeep] No, it’d be my pleasure. Thanks once more, Ryan. Thanks listeners.
– [Ryan] All proper everybody, thanks once more for watching that episode of the IoT for All Podcast. Should you loved the episode, please click on the thumbs up button, subscribe to our channel and make sure to hit the bell notifications so that you get the most recent episodes as quickly as they turn out to be accessible. Aside from that, thanks once more for watching and we’ll see you subsequent time.
