Custom Automation For Your Business

ABOUT THIS EPISODE

Do you remember when George Jetson pushed a button and a salad came out of a box? Believe it or not, that future might be closer than you think. 

Born out of demand during the COVID pandemic to automate and digitize food supply, this manufacturing company has created an automated food delivery system that sounds like it belongs in a science fiction film. But it’s not just that it’s as fast as you think it is. Abram Simon, VP of Software Engineering at Hyphen, shares not only how close this technology is to becoming a staple of commercial kitchens, but also what it will do for order accuracy, labor flexibility, and what comes next.

Join us as we discuss:

  • Plate appeal - by a robot?
  • Automation and labor flexibility
  • IOT solutions for lowering the entry barrier

You are listening to application modernization, a show that spotlights the forward thinking leaders of Highgro software companies. From scaling applications and accelerating time to market to avoiding expensive license and costs, we discuss how you can innovate with new technology and forward thinking processes and save some cash in the process. Let's get into it. Thanks for listening to the application modernization podcasts presented by shadows soft. I'm your host, Nick Markarelli. Today we spoke to Abraham Simon, is the VP of software engineering at which I'll allow him to describe, is in the restaurant automation space. So we had a great conversation talking about how kind of this new, very emerging portion of tech is being applied towards taking fast casual restaurants and making them more efficient for the consumer, more efficient for the business owners and, you know, lightly touching on the technologies involved and how they approach and how they do things. So really interesting conversation if you're interested in food, like I am, or if you're interested in how some of this new tech is being, you know, brought into the consumer space, I think you'll find this great. Thanks to our sponsor red hat for continuing to sponsor the PODCAST. Enjoy the conversation with they. Hey, arm thanks for joining the podcast today. How we doing great? Thanks for having me. Appreciate it. Super excited to have this conversation. So, you know, we always like to start the podcast we want to know who we're talking to, what they're all about, where they come from. So give us a overview of you know, your experience, maybe your journey to where you are now and everything that would be interesting for our listeners to understand about you. Awesome. So, yeah, I've been doing software for about twenty three years now, which is man, it's a long time. It's been a journey. So most recently I found myself at where on the VP of software engineering, leading up a small but growing team of folks working on restaurant automation, restaurant operations more namely it's an area that's extremely passionate for me. I started off my my professional career in my teens as a cook at a small sort of neighborhood restaurant. Did lots of things food service related, from cooking to doing dishes, delivering pizzas, whatever you can think of. That culminated in the early thuds with opening up a coffee shop. So I owned an operated a coffee shop that you know. We made smoothies, coffee, sandwiches, soups, all kinds of great stuff. And Yeah, culminating all my experience there with restaurants to operate one for a few years, which was a ton of fun. It was one of the best times of my life. But ultimately I couldn't get away from technology and found myself back in software and yeah, come into today, where I had the chance to join and put all of that passion and experience in play. Combining food service and and software together. It's a dream job for me. Gotcha, that's cool. So obviously the next the next question I would have is, you know, tell us a little bit more about and how you serve the market and your customers and what type of use cases you are you solving for. I know that when I started looking to I got really excited because I'm also a fan of food and why I like to eat, but um, but also like to cook. So you know, the concept of what your restaurant automation is doing is...

...feels like a wave of the future. So you would do a better job explaining than me. So tell us a little bit about hyphen like full advertisement for what you guys are doing. Absolutely, it is definitely the wave of the future. was born out of the pandemic, more or less. We have a pretty good pedigree within the company, a lot of people who've worked in food service before, be at instagram or cafe x or any number of of either traditional food service establishments or automation type of companies. So we have a big passion for it and we saw the trend coming coming through the pandemic, where a large shift, almost a forced shift, was happening to the digital ordering ecosystem and many restaurants who had had struggled with very thin margins couldn't make it through. In the ones that did were able to do so by embracing automation, embracing the digital ordering wave of the future, and have shifted significant portions of their sales to that, to the point where they're looking to the future of having fully autonomous type of operations where they don't have a dining room, they don't have the traditional takeout service or eat in service that they might have otherwise had, but going for purely digital type of storefronts. And of course that's most common in the the QSR space, the quick serve restaurants, the Chipotleais and different salad and healthy food type of restaurants. Fast, fast salads is a big trend right now that we're latched onto. So that's that's our core spaces, automating the production of those foods and ultimately through the the operations, all operations within the kitchen. That starts with our digital macline that we've created to the hardware product that we deploy to to restaurant and it's combined between Backu House in front of House. So some just have a replacing the existing mainlines they might have in their kitchens. Others put it front and center so they can can serve customers right there. The ultimate goal is an order comes in, you place an order from saying Uber eats or a door dash on your phone and finds its way through all the pipes of the internet down to the machine. It's able to produce that southing. It's fully loaded with ingredients, it's able to produce that bowl fully autonomously pop it out the other end ready to be delivered to the customer. The way of the future. It's here. The yes that so I've so many questions. Here they come. So what I've seen, you know from your website is, you know, there's a probably a live image, I guess, or maybe it's a mock, but a lot. You know, this incredible counter preparation area that's got doors with all these machines. I assume that's where we're all the food is stored and like, how did you guys? How did you guys get there? Like, how did you go? How we're going to build this piece of hardware was I mean, you know, programming hardware is different than just building software, so I'm sure this had some type you know, you're looking for certain types of team members, you're looking for certain types of skills. How hard was that to make that come to fruition, I guess, is what I'm what I'm getting at? Yeah, it is. elegantly. It's actually been super fun putting together a team of individuals to work on this because it does require a really diverse set of skills. But, more than anything, we hire for non traditional backgrounds. We really care about being empathetic towards our customers and understanding the challenges that restaurant space and the challenge the...

...consumer space to and they're ordering food and so having folks such as myself who were born through food service and came up at that and have lived the life and see all the troubles that that come along with both operating restaurants and working the line, just being part of the restaurants themselves. So we've optimized for that a very diverse skill set. Getting generalists who can come in and solve problems is the number one thing. Of course it's been challenging with remote work in the in the days of the pandemic and building a team up remote first has been fun and challenging at the same time. Sure, yeah, we've from everything just from software to operations to hardware engineering. Everything is just we're full of people who care, people who are pouring their harden souls into this and having a lot of fun solving really challenging problems. So does your does your offering, which you know, num referencing what we've been talking around about right. So your your offering? Does it come in multiple sizes? is a custom built for each customer? Do you guys have like you know, this is it comes at twenty feet or yeah, whatever that might be. Yeah, the so the biggest differentiator for us in our robotics platform is that it's not a standard robotic arm that you might find. When you think about automation, we think about robots. You think about humanoid type robots or giant arms that, right, can do very generic tasks. The problem with generic, sort of all encompassing automation like that is that it's not purpose built, it's not efficient at what it's doing. So we've built our platform from the ground up a number one to be a drop in replacement for what they already have. Steven, our CEO, likes to use the escalator example, where if you have an escalator and it breaks, you end up with staircase. So no real harm done. We took the same philosophy and building out our machine. It's the same form and function as a makeline that you have within a restaurant already. You've seen them if you've gone to a chipotle or a subway or a salad bar anywhere where you have all those little hoppers full of ingredients. Right. We fit the same form and function. So we're counterdepth, counter height and as much as twenty five feet or even more long in the Makeline, something that people in a kitchen would be very familiar with operating. The difference is underneath the surface is a fully robotic platform that takes care of everything for you. So it is customizable. It's built in a modular way where we can tune in. We have a really great culinary team with an executive chef who hounds in on every ingredient on the menu for our customers and make sure that we're able to dispense that in very high quality ways, that we don't destroy things, you know, destroy the food and end up with soup when you wanted to sell it. So it's precise, it is high quality and it's fast, much faster than what they have today. So it's custom built of the modules that we support and tuned in for each ingredient on the menu and then delivered custom for those restaurants. Gotcha. That's really cool. For those listening Usecom you need to go see this. This is very cool. Yeah, we have some great videos and whatnot. A bit in action up there. Is there a asking us this on the spot? So the answer can obviously be knows. Is there. Is there a customer story you can share? You don't have to name the customer, but just now, not not yet. We are. We're actually weeks away from deploying our our very first line. I believe that'll be up in the Seattle area.

It's so maybe I'll revisit this in a soil follow up podcast story. Yes, I like it. How does this impact the traditional restaurant workforce? Obviously with some you know, we through makeline automation. You know today there's people doing that work, for better for worst, probably for ours. I never, never get my double chicken at Chipotlea not a sponsor with the podcast. So come get me, come get me, Chipola, but you know, I assume I see him. That's one of the things that makes this type of restaurant automation really great for the consumer is that when you order a double chicken and no beans or whatever that is, it probably happens with more accuracy than if I just go in and put a web order and there's a human doing it. But that also changes the way a restaurant looks. So what are some of those changes that you would see as someone who worked in the industry? Is worked in both industries, right at the software side and the restaurant industry. How do you see that changing over time, because restaurant automation is not not a new idea. Lots of people have been talking about it. We just haven't seen many people execute on it yet. You know, we know a lot of the you know large chain fast food restaurants have been talking about this for years. I know there's some like demo stores and place around these concepts. How do you think that changes for the industry as far as food goes? Yeah, one of the biggest impacts you covered it for the consumer. Right, they get good portion control, so they get what they've paid for. They get what they asked for. They don't have to worry about you know, humans make mistakes, so they don't have to worry about that and get cheated on something that they asked for and didn't quite get. Or a mistake is made. The just impacts the recipe overall and it's it's not what they wanted rights, not even quantity, the quality control there too. So that's a big factor for the consumers as well as from the consumer side there's just been a huge shift towards digital ordering and if you're ordering something digital but there's still a large human component to it, then you're the experience is reduced. It's not quite as robust of an experience and, of course, long tail, you know this will get into autonomous driving and delivery and you know, finishing out the last mile. They're they can all be autonomous, but right now we're focused on the kitchen operations themselves. So from a Labor standpoint, from the folks working in the line and the restaurant they they have a problem with elasticity of labor. And so you know restaurants are busiest at lunch time and at dinner there's usually a two to three hour rush where you're super busy for those two three hours and then maybe there's not much to do afterward. And so you don't want people don't want to be working just two or three hour shifts. That doesn't make sense, that doesn't pay the bills and they're not going to get really good at what they're doing that way either. And so what this allows is for people to have higher level impact within the organization. They can either do that from a hospitality standpoint and do more customer interaction when they do need to be interacting with customers, or they can be operating the the machines. They can be doing higher level technical work or or less repetitive work to to spend their time rather than spending it just picking an ingredient and put in the bowl over and over again. So it's it's very beneficial from a job satisfaction standpoint that it gives them more control over the timing of when they're needed and how they're utilized during those times and gives them a little more autonomy as well to be doing things that are more impactful and more important. So kind of off the top of my head,...

I would think if I'm a LEU'd say I'm a local fast casual restaurant and I need to staff, you know five, five to six people in the makeline. Then I've got, you know, people doing batch cooking and then someone on the register and maybe somebody floating. I mean that's, you know, ten people, some type of manager probably with with an automated bank line, I could probably invest in a higher quality collection of line cooks or chefs instead of the Makeline, because the makeline so probably just change. I'm thinking it changes the dynamic of what my restaurant could be opposed to what it has to be, because I deep hands, I'm replacing hands, but I'm also could be lifting the quality of maybe my food and my recipes and customer sad you know, as you pointed out exactly, yeah, we're we're not cooking the food. So we're great at assembling food into a finished product, but we're not aiming to cook the food or prepare it in that sense. So a lot of these restaurants they build their brands off of their their culinary aspects right, of the recipes that they have and the creativity that goes into those, and then they invest a lot in in the preparation that goes into those. Of course they're also worried about how the final presentation of it, but that's something that our machines can handle pretty easily. So what what our machine does is, yeah, it allows you to refocus your Labor into places that are more impactful, but it also enables the other functions in the preparation, the cooks and and other prep duties to be more informed, because our machine right now there's kind of a black hole of information. When the inventory comes in, it gets prepared, it gets turned into meals and then delivered, but how much was actually used? How much of it was wasted to shrinkage or other right for reasons or how often does the kitchen get backed up because they weren't prepared adequately for the rush that was about to come in? Or if the rush last longer than you would have expected, maybe they just didn't have enough prepped for it? Because we have, down to the in downd of the gram, level ingredient information about what we've used and what's available, we can let the cooks know ahead of time that they're about to run out of chicken and they need to start cooking more and get it prepped up so that the machine is never starved, whereas right now you'll hear, if you walk into one of these restaurants, you'll often hear somebody on the front line shouting we need more chicken right and you know that that's usually too late by the time they've called that out. So utilizing all the data that we can collect off of this also enables all the other functions within the kitchen to run more smoothly. And then, I guess if you you plug in the you know the digital ordering world that we live in now, you have twenty five orders come in for, you know, beef Kebab, then you know immediately. While I don't have that much in the sheet right now, in the machine right now. She was thinking about something else, I guess. So then you, then you, then your cooks know. Well, we need to we need to get some kebabs and roll on. So I guess you know. If yeah, I mean that makes the customer experience uniquely pleasurable, I would imagine. So yeah, absolutely. And also tying in the middle part. So right now you know if you ordered from a newber eats, they'll tell you an estimated delivery time and they'll be able to give you some good tracking as to whether it's been sent to the restaurant or not and whether it's been picked up by a courier for delivery. But there's a big...

...gap right there in the middle when it actually is going through production and assembly. And so with our machine being fully cloud connected and, you know, connected all the apis of the different providers and whatnot, gives us that capability of knowing what has already been done, where in the process are the Bulls that you've ordered and how close to finish it is, so we can give better feedback to the consumer about the timing mechanisms and also the machine being a much higher through put than a human can achieve, especially sustainably. Over time, we can give much more accurate estimates in the kitchen as to when those things will actually arrive to the consumer. Right now, a lot of these restaurants create buckets of time. So they might say we only take, you know, five orders within this fifteen minute interval because we know that's what our production capacity is. And so they'll force the consumers to look forward in time and say, okay, I'm going to to this forty five minutes from now for pickup perhaps, whereas we can, we can narrow that and we can say we don't necessarily have a limit in the amount of orders that can committed a specific time because our capacity is roughly but for x throughput of an average staffed line. So we know we can produce a much higher volumes. And it's also automatic, elastic capacity. So if the machine is city idol, that's fine. A large catering order comes in, let's say, and the machine can automatically kick into ear and handle that capacity as supposed to calling around to your staff and saying can you commit on short notice, like we really need somebody. Somebody called in sick today and we just don't have capacity to fill this and it's a cascading effect that the whole kitchen gets backed up because of that. Right. Interesting is there? Is there a thought about other food service kind of verticals such as like, I think, cafeteria hospital? Is that something you guys have discussed and I guess you wouldn't really have to design or on it because your product is like a kitchen utility, like your kind of just fits wherever that is. But is that, is that an area of emphasis maybe at some point? Yeah, absolutely, we already have. Some of our reserve units are already taken by copackers. COPACKING is a pretty it's the industry you're describing right there, right so right they'll produce larger quantities of sort of out of the box recipes that they'll deliver to the universities or airports or hospitals or those kind of things. So we can be used in a not industrial setting in that sense. So it's not, you know, large scale manufacturing of food products, but sort of that middle tier that's somewhat underserved right now. We can absolutely serve them as well as if people want stuff, you know, those are those tend to be fresh. In the twenty four hour range, so they might make it in it gets consumed within, say, twenty four or forty eight hours, as opposed to the you know, immediate impact that you get from the QSR space, where you know it's consumed within, you know, thirty minutes or so of production. So yeah, we can, we can service both of those spaces very effectively. Got It? Okay, cool coming up with ideas that you guys have already thought about. You'd mentioned that Y'all's technology can take the let's call the plating or the the way, the way the food is meant to come out. Now it's mental. Look, how do you do that at a high level? There's probably some secret sauce, but I just I would just love to understand conceptually what your makeline automation does around that. Yeah, so a...

...kind of opposing it to what you might think of with a robotic arm, where it'd be picking the ingredient and moving those to the bowls. Instead, our bowl, the ingredients are all stationary within the makeline and the bowl traverses a conveyor on the on the bottom where ingredients are dropped into it. So as it's traversing that conveyor has the opportunity to arrange the bowl properly by rotating it. And so when we do that we can take into account a lot of interesting facets where we might want to put a specific ingredient on one side of the bowl for the plating requirements, for the visual accuracy that the restaurants want, but we might also figure out that the density of this particular ingredient in the weight of it is going to cause a load imbalance, it's going to put off the center of gravity of the bowl, and so we can make intelligent choices in real time based on whether you've ordered extra of a heavy ingredient or you've asked to omit a certain ingredient. We can actually change the rotation in real time to make sure that the center of gravity stays in place, that the ingredients fall in the correct way so that the arrangement is not only appealing but is easy to mix for the for the consumer. We can take all that into consideration to make sure that we have the best placement of each ingredient within the bowl. In addition to that, we take photos and have videos of the Bulls going through the machine. So quality control is a big problem. You know today, if you go on Uber Eats and you order something and then you say I don't like it, it wasn't high enough quality for me. They take the consumers opinion. They're at face value, and so that's a cost to the restaurant that they might have made a very high quality product, but the consumer is given the first pass on that one. Right, let's defer to them, and so they could be losing a decent chunk of money and that, you know, impacts their their razor thin profits when they have to do that. And so it the plating of it in the accuracy of the ingredients, having those photos with, you know, machine vision models and stuff on top of it, to specifically call out the accuracy in the quality of what we've done and empowering the restaurants to to appeal those decisions with the delivery providers gives them a lot of power to control their profits. In that sense, now that makes sense. One time I got a chicken Burger. It's not what I ordered, so I let Uber Eats. No, I didn't order a chicken Burger. The Chicken Burger was good. I did, but I was I was sad, for I mean I'm not a guy who orders a burger and hopes to eat chicken, but it was still pretty good. There's no replacement Burger and not at all. So you know, you're you're operating in a pretty pretty emerging space. So you're focused on these things every day, it seems. So what's something that's maybe a little further out, just intact? It doesn't have to be specific to restaurant automation, but, as a technologist, something that interests you that's, you know, kind of creeping up and is starting to might be an interesting trend. Whether it works or not doesn't matter, but something you find interesting seeing around the corners, what I usually like to call it. Yeah, and you know, I I don't know if it's just my my exposure to it recently and maybe it's been around for a while, but the evolution of the IOT space, his his really come along in the last few years from from what I've seen, and you know, where it used to be that you had to home grow solutions and kind of figure out all the pieces yourself,...

...the major cloud providers are providing really amazing tool sets for just really reducing the barrier to entry on developing, I secure IOT solutions and you know, cloud flare recently announced that they were going to use MTTT as their protocol of choice for all of their pub sub distributed systems, which it just leverages, you know, the core of Iot and the heroistics involved there. So the investment coming out of these clog providers that sort of democratizes access to them, combined with the rise of, you know, the arduinos and and Beagle bones and raspberry pies of the world, it really creates this hybrid ecosystem that used to be off limits to to, you know, a small hobbyist just exploring what they can do at this so it's been a really interesting see change there, to see so much investment in opportunity come out of it. Yeah, that I mean the whole it's funny. It wasn't that long ago everyone was talking about raspberry pies and they're like hey, there are these new cheap things that you can do stuff on. And you know, not not being an engineer by trade, I was like that looks like a PCI cards. All right, you know, I was just like whatever, but then I then you know, the evolution of that is, you know, people are building, you know, small, small systems that they're putting everywhere and you know, even our engineering team here, you know, took a bunch of raspberry pies and kind of plug them into a display and created this, you know, the's own little node for like our knock and it was like, Oh, oh, that's how that's used. I think sometimes I consumers like myself, you know, I'm sales technical, not technical technical. Go I don't get that until we actually see an action. So that's that's what's kind of cool about what you're doing. It is like I can see that, like I go to your website, going to go, oh, that's what all that technology is doing. Yeah, it's very interesting to I mean you look at the the initial use case, like you're saying, and it's like, well, what am I going to do? Light up some led's or, you know, display something on an oled and it's it's cool, but it's not too functional. It's not something that adds a lot of value in your everyday life. But, as has been true for you know, a decade or more now, data is power and having these powerful edge devices and a robust ecosystem in the cloud that they can be connected to to leverage that data, having more processing power at the edge where you can run machine learning models or machine vision on the edge and then take the aggregated information that you get out of that and ship it off to the cloud for further analysis. Is just that the ecosystem is a whole. It's no longer individual parts that are somewhat novel, but it's really an ecosystem at this point that if you have an idea, you can can bring it to fruition. What's some what's something you've learned of? It could be lightly over the course of your career, something invaluable advice, a lesson, learn that kind of thing. That would be interesting to our listener base. You know, I don't know how obvious it is. It seems obvious to me in retrospect, but it's something that I feel like I continue to learn every day, which is just speed, and you know the old saying of failing fast. Having speed be on your side when it comes to creating products is one of the most valuable things that you can concentrate on. And you know, our team is hyper focused on making the smallest changes possible as rapidly as possible, and it's all about the feedback loop. There's, you know, a lot of engineers like to sort...

...of hunker down and and get something perfect over the course of weeks or months. And if you made one wrong assumption at the beginning and you didn't work in public for it, it's probably going to need some course correction. And if you took too long and you didn't work with your team and iterate fast on it and set up the tooling around the engineers to make sure that they can iterate really fast and really safely, you're going to make a lot, a lot more mistakes and you're going to pay for those mistakes in a much larger way. So the biggest thing I have is it honestly, it doesn't matter what technology you're using, what language you have, whatever the newest framework is out there. The thing that matters is that you're making choices that lead to very rapid iteration and lots of speed introduced into the process. That's it's a great point. I think there's a sometimes intach we hear. We hear about speed. You know, we take take, you know, Netflix, for example, and they, you know, they've built a brand on speed, whether it's how they develop software or how much content that they collect and get out in the market place. To you know, really grab consumers and it feels a little too far fetched, you know, like Oh, Netflix as is brilliant engineering team. But you know, I think that actually applies to even non technical things. So we had a training academy for years at Chat US offt where we had people come in and in turn and we would teach them how to do the stuff we do. And one of the one of the interns, was trying to skill up to pass a certification. He said, I got I know all the answers, it just takes me longer than the time allowed. And I had, you know, I think, I think a moment of genius for me. I said, well, you know, speed is a measure of proficiency and I was like I should write that down somehow. That's a good idea. So, you know, I was, you know, explaining. I'm like, if you can't do it fast, you don't know it, like you got to really know it. So what you're talking about there, I think, is interesting when it comes to, you know, developing software, building platforms, whatever it is, if you can do it quickly and then you can even the skill of getting feedback, like lots of organization struggle with that today. You know, everyone's talking about agile develops and all those are nice words, but the end of the day, if you can't, if you can't rapidly prototype whatever you're doing and then build upon that, it's going to take a long time, and that's it's a cultural thing as well as a skills and proficiency thing. Absolutely, and yeah, it's great that you brought up Netflix, because they're actually very inspiring in this sense. They relatively recently, within the last I believe, twelve months, reorganize some of their teams and have a team now focused on engineering productivity. And you hear productivity, it's kind of a dirty word people, you know, don't like the bought coming down on them saying they need to be more productive. But the the the root of that team. In the same for our platform team that that I'm building up here, at the core of it is to remove friction and, as they've stated it, to pave the roads. And so having admitting that a lack of productivity or a lack of speed is not fundamentally attributable to the too specific engineer, it's not necessarily that they're they're not working hard enough or they're not smart enough to solve the problems, but rather setting up the platforms and processes that encourage them to iterate really fast on things and to have that safety net that I was talking about before, that it's okay to make some mistakes. It's okay to...

...roll out a feature that's half baked and get some feedback on it rather than try and solve all the problems yourself and, you know, go into a cave for for a while. Just having that culture, like you described it, to to embrace quickness of things. You know done is better than perfect in that sense. So right, yeah, just just embracing it at a cultural level. And then it gives you an interesting perspective when you look at the tools, like you mentioned, like agile as a methodology. That's really a tool. That is not the goal in and of itself. It I don't care if we're agile or not. What I want to know is, are we getting the benefits out of being agile? And that gives us the opportunity to drop some of the ceremonies that don't really matter to us or don't help us with the end goal, but embrace the things that do actually help with the goal, and the goal is ultimately to move fast, because that gives us great products, that gives us that feedback loop. Yeah, I was just reminded the you know, when your reference Naturalix, you can talk about, you know, three or four major concepts and technology. Some of them you know, wid spread, some of them not. It made me think about, I don't know if you remember this trend a few years ago, the whole concept of chaos engineering, MMM, which, you know, maybe in you know, in the type of offering that you're building and IFE and makes a lot a sense. But I remember when, you know, that became popular and there was all these kind of commercialized offerings around. Hey, we can help you do chaos engineering, you know, for your organization, and I thought it was really interesting that people would build a company around the concept of, you know, Netflix used to go, reliabilities the most important part of our business, so we need to be testing to make sure that our systems are robust and reliable. They had a great use case for it. We need to try to break stuff to make sure that we can fix stuff fast or it'll self heal or whatever. That is right. So then then it became a commercialized thing. We're like hey, you know, we're going to go sell this to, you know, a fortune, one hundred bank they they love chaos no, they don't. They don't want chaos engineering. It's not. They want uptime. Right. So it's kind of funny that there's degrees of how somebody would want something like that. But you know, Netflix is you know obviously Amazon and facebook. They're all building these things out of the art of the possible, and then we have to take the Little Pie bits and pieces of that and go does that apply to our use case? Chaos engineering for you might actually be really interesting because you want something operational, you know. So it's scale. That might be something you look at. Maybe. I don't know absolutely, and you know I look at these. Again, these are tools and understanding the fundamental goals and nature of the tools and how they apply to your problem space in your team and creating the culture around it is so important, which is why you can't just bottle up that magic and sell it to another company. Right. It takes work to understand how to integrate it into your system. and to me chaos engineering is fundamentally about the fact that the as a distributed system grows in size, it gets complex enough that no one engineer can understand the cascading nature of failures within the system. Right. It's the root cause analysis. It's never just one thing, it's a cascading failure of many things. And so, you know, just like no one person knows how to build a toaster or make a pencil anymore because the world has gotten so complex in the supply chain and whatnot, the software systems that we build are sufficiently complex that no one person understands the ramification of changes that they're making. And so introducing that chaos...

...engineering, so that it exercises the unknowns overlooked parts of your system and identifies those cascading failures, gives you the opportunity to put in circuit breakers and bulkheads and other mechanisms to prevent that into ensure that they're continuously implemented and they don't become a side thought that only gets thought about once disaster has happened. So, yeah, superminutes, that's a that's a great point. You know, it's all about you know, like at what you said there. It's about your problem space. So sometimes we're taking broad techno technology concepts and going, you know, just use the public cloud. That I mean that's been one for a decade. You know, like sometimes that doesn't make sense. I mean, I think a lot of times it makes sense, but depends what you're doing. So yeah, yeah, and for our use case in particular, for chaos engineering, is is especially important because if a if a restaurant loses connectivity to the Internet, it they have to still be able to make food. They can't shut down a restaurant because the innerent Internet connection went out. And so leveraging all of the benefits that come with having a cloud connected machine or fleet of machines, but also having the failovers where, like the analogy I gave before about the escalator, it does turn into a staircase when it breaks down, right, and it doesn't turn into a slide that just ejects you offer off the platform that you're going for. Same thing for our machine. You know, worstcase scenario if you're not connected to the cloud, your digital ordering platform is down, if your Internet connections down. So we have to have a way that they can, you know, print a receipt with the order on it, manually input it into the machine and keep the machine rolling. Even in the worst case scenario, right there's a power outage, they can lift up the lids. They can start pulling out ingredients manually and still making their bowls just like they did before or we arrived, so they get all the benefit and hopefully none of the downside. Now that works. I mean that it's a whole of the value problem works out. I'll like it. Lairy, thank you so much for the time. It's a really great conversation, very interesting to me. I know our listeners will really like it. So thanks again for carving out an hour for us to talk about the new wave of the future and, you know, restaurant and delivery and all those fun things we're rooting for. You think it's really cool. Absolutely. Thanks for having me. And Yeah, one last Shandele's plug here. I'm higher and engineers. I need I need great people of all disciplines. If you're interested, reach out, let me know. Awesome. Thanks so much, and folks, go look them up. It's cool, and go get a job with them. Sounds Great. All right, thank you, sir. Appreciate it. Thanks. TACARE. Application modernization is sponsored by Red Hat, the world's leading provider of enterprise open source solutions, including high performing Linux, cloud container and couper Neetti's technologies. Thanks for listening to application modernization, a podcast for high growth software companies. Don't forget to subscribe to the show on your favorite podcast player so you never miss an episode, and, if you use apple podcasts, do us a favor and leave a click writing by tapping this ours. Join US on the next episode to learn more about modernizing your infrastructure and applications for growth. Until next time,.

In-Stream Audio Search

NEW

Search across all episodes within this podcast

Episodes (25)