HH81-13-05-22.mp3

Harpreet: [00:00:09] What's up, everybody? Welcome and welcome to the artists and data science. Happy hour. It is Friday, May 13th, 2022. I'm super excited you guys are here. It's been a long time since I kicked off the happy hour as we host shout out to Greg Cook for taking care of last week's episode. Great job. Really appreciate you taking over the ones and tools for me and for anybody else who's been able to step up for me throughout the year. My guys are awesome. We appreciate you all. Also, huge shout out to our sponsor for today. It's going to be the MLW Ops World Machine Learning and Production Conference. If you guys have not heard of this conference, definitely go and check this out. It's a great, great lineup. You've got speakers there from meta feast hugging face, DoorDash, Yelp, Yahoo, Google, Spotify, Shopify, eBay and so much more is taking place in Toronto, is taking place June 9th, 10th and 11th. I will be there and I hope you are there as well. I'll be hanging out not only at the package booth, but I'll be demoing the package and product. Just create the demo labs that you are kind to me and don't mess up my live demo while I'm out there doing it. Yeah. Huge shout out to the MLPs World Conference. Thank you so much for being the sponsors of the Data Science Podcast.

Harpreet: [00:01:38] If you're just a data scientist and you don't see an ML ops event as something that's for you, then I think you might be wrong. Because not only are we going to talk about ML Ops at this conference, we're going to spend topics ranging from AI ethics to model governance to team building strategy, and there's also an entire track dedicated [00:02:00] to the business side of machine learning. These talks will leave you with golden nuggets of wisdom that will fast track your ability to build a data science practice with scalability in mind. So definitely go ahead and check this out. There will be a link in the show notes to the conference, and if you're in the room here, I'll send a link to the conference right here in the chat. Use the discount code HARPREET to the capital H for 15% off your ticket. And if you do. Com. Please do not be shy. Come in. Say hi to the pachyderm the entire time. Also. Released an episode today on the podcast with the one and only Dave Langer. Dave Langer, where have you been? This. This guy. That's awesome. It's been quite a while since we had Dave on the show, but when I was first kicking off, he was an integral part of just making it a great session. So Dave, thank you not only for coming onto the podcast as a happy hour participant, but finally coming onto the show as a guest.

Harpreet: [00:03:06] So thank you for that. There's a bunch of other episodes released over the last few weeks that you should check out if you have not already spoke to Marcus du Toit, Oxford University Professor of Mathematics. You might recognize him from a bunch of different shows on the BBC, the history of Math Being One. He's also written a couple of New York Times bestselling books. One of my all time favorites is The Creativity Code, and this is a book about essentially how deep learning can augment human creativity. Amazing. Definitely check it out. I've also, a couple of weeks prior, released an episode with the people data scientist himself, the one and only Sara. That was a great conversation. I hope you guys check that out. I also did an episode with the data Professor Smart. That was a couple weeks ago, so go check that out. Also, Christine was on the show, so go and check that out. And [00:04:00] I also did a non data science participant, essentially an episode with Natalie Nix and Dr. Natalie Nixon. We talked all about creativity and how you can be more creative. So definitely great episodes. Go and check those out. Apologize for any noise you're hearing about part of town here. So, look, we're going to kick off the conversation now.

Harpreet: [00:04:21] Here's here's what we want to talk about. Everybody always talks about how 80 something percent of data science initiatives fail. All right, cool. That happens. Shit happens, right? But what about that 20% that actually do make it right so that 20% of data science initiatives that do succeed? What is so unique about that? And right off the bat, I want not obvious answer, the only right answers like there was alignment between the data science team and business stakeholders, or they try to go the overly complex model. Those are obvious reasons why things would fail. What are some non-obvious reasons things would fail? Whether it's systemic issues of business, whether it's data science, behavior or misbehavior. What are some non-obvious reasons that a data science initiative would fail? Let's go through this question. But then after this question, just on spotlight here, I do want to talk about recent posts about the layoffs. And there's been a huge round of layoffs everywhere. Should we be worried about data science as data scientists about this and what's going to happen to protect the economy going forward? I didn't want to talk to you about that in a little bit then, but if you don't mind, I'd like to go to you first for this. What is it? What's so special? What's so special about these 20% of these initiatives that actually succeed?

Speaker2: [00:05:49] Yes. Thank you. I'm so happy to be here. Can you hear me?

Harpreet: [00:05:54] Okay. Okay. Loud and clear.

Speaker2: [00:05:55] Yep. So happy to be here. This is my first time I've been attempting [00:06:00] to make this so. So for me, I think 20%. In my experience, what has worked is the change management, right? So sometimes you have a solution, but in a certain situations, almost majority of situation was something you would bring there. The change in the process is sponsorship. How do you bring people together to adopt that? So adoption is really the key in my scenario and there have been situations where the posse looks good and after that you implement it. They are not many takers. Right. So I can I can tell you that having that change management to be part of that. Also, we are technical people. And one thing I have I've noticed that what we think usually we like but usually is totally different. Right. You think this a great solution of a building you put in front of them. So yeah, it looks good, but it's okay. If it looks good, let's go and implement it. And what we see there is that there's nobody using it, right? So there's a whole element of that is that how do you really understand what users want? What do they experience for them? Right. And so some of the reasons I have seen those things really work well if you bring the user along with that.

Speaker2: [00:07:19] That in a journey of whatever the solution we are basically developing, right? So I would say the change management process that comes with is sponsorship. As things are changing, you have someone to basically make that happen. Users are part of the journey and of course you are addressing the real problem, right? So so think about why do somebody would care about my model. You know, it might be a business problem, but if you go or set up a group of people, is there is do people really care about that? Have we have we really taken care of those those needs that they are basically interested in? Right. Another thing that I can see that very [00:08:00] successful is that and sometimes a problem never happens in isolation, right? So there's somebody upstream causing the problem and somebody downstream getting impacted. So when you look at the solution we are trying to solve, you want to bring those people as well part of the solution. So the impact is bigger. All of a sudden there are a lot of people willing to adapt and making the contribution. So let me stop there. I think those are some of the things I've seen have worked really well, at least in my experience.

Harpreet: [00:08:34] Thank you so much and I appreciate the insight. Let's go to Mexico next. And then after Mexico, let's go to Jean and then Tom also shout out to everybody else that's hanging out in the room. Good to see all you guys here. Sometimes you do take it. See again, Russell Willis is in the building at the time that Avery Smith has just entered the building. Good to have you, Avery. Also Thomas as well. Thanks so much. We're going to hear from you. We'll go to Ben and then we'll go to Tom.

Speaker3: [00:09:05] Yeah. I guess I feel like the reason why some of these projects like succeed and I know we said non obvious answers but I don't know, maybe this is not obvious or it is obvious.

Harpreet: [00:09:16] Sometimes it's like the people on the project.

Speaker3: [00:09:21] Actually maybe that's mean answer, but yeah, like sometimes it is right. Sometimes you just don't have the skills capabilities yet to make a model in production work. So I think that is a really real one. I think a second one also and I wonder about this because when we think about like the ML engineering role, sometimes it's kind of like it feels like it's there to make up for a lack of investment in infrastructure and infrastructure teams.

Harpreet: [00:09:52] Because when I think about it.

Speaker3: [00:09:55] I kind of feel like if you treat models as sort of D composable [00:10:00] or kind of modular assets within like a product or service or platform, hypothetically, whatever model you work.

Harpreet: [00:10:07] On shouldn't.

Speaker3: [00:10:09] Matter.

Harpreet: [00:10:10] Like from a.

Speaker3: [00:10:11] Will it succeed or won't it succeed? So in.

Harpreet: [00:10:16] That regard, like some companies, I.

Speaker3: [00:10:18] Feel like I feel like they have ML engineers because they essentially don't want to put effort into infrastructure. What I see in like really successful teams sometimes is you even don't have ML engineers, you have scientists, and then you have like the MLPs or like data engineers or data ops teams who kind of work on stuff. And then essentially the data scientist kind of kick off a model or like, you know, they deploy the pipeline or whatever. And then there is like this system for which the model then goes down the cattle chute to then be like eaten or something enjoyed.

Harpreet: [00:10:52] So yeah. Thanks so much. We appreciate that. Let's hear from then, by the way. Anyone listening either on YouTube, LinkedIn or even here in the chat? If you have any questions at all, please do not hesitate to drop the question right in the chat. I will add it to the queue and we will get you some answers. Let's let's hear from then again. Just the question we talk about is that non obvious reasons why only 20% of data science initiatives actually succeed non-obvious answers only. So nothing about overcomplicated models, nothing about no alignment between business and data science, just completely non-obvious answers. Let's go to them and maybe if you want to chime in, let me know. Russell, if you guys want to chime in, let me know, raise a hand or whatever. And then Tom.

Speaker4: [00:11:52] Yeah. I think there's been a couple of studies that I've seen where they asked companies with data science teams, you know, have you put [00:12:00] anything in production yet? And somewhere between 70 and 80% are like, what? So that's one. One reason is and I think Vijay and Akiko both hit on this like they can't get it into production. It's kind of crazy to think that there are that many companies out there that have data science teams and they just they've never made it that that last mile. I think I've heard people call it the last mile problem in AI. And so that's one huge reason. The other one that I see so often is you've got to prepare users for models. It doesn't work like software does, software is. It works to do the same thing. At least when I code it, it works. It'll do the same thing every single time. And as long as it stays within the specification of what it's supposed to do, it'll work. But models.

Harpreet: [00:12:55] Don't.

Speaker4: [00:12:56] Models will behave erratically, and you can ask models to do stuff that you can't. I mean, think about Alexa. If you ask Alexa a question like the weather, she's always going to give you a good answer. But, you know, you don't have a specification for what you can ask Alexa. You could ask her a ton of stuff, but you start getting to the edges, like, where can I find green jellybeans? Maybe she gets you an answer. Maybe she doesn't. Maybe she goes, I have no idea what you're talking about. So here's a picture of one. But you know, and so and that's the cool thing about models is that you never know. You might find some new ways to work with them, new ways to collaborate. And I mean, human machine teaming is actually becoming a thing. But if you don't tell users like this will not work every single time the way you expect it to, they get it. It doesn't work. They stop trusting it. They get rid of it.

Harpreet: [00:13:49] Thank you very much. Alexa. Where to answer it? Anywhere other than the Jelly Belly factory. I'd be very, very upset as a jelly [00:14:00] belly factory is right down the street right there in the layout. Shout out to Tom. I hear from you. And if anybody else wants to chime in on this question, by all means, just raise your hand. Let me know the chat and we'll talk over it.

Speaker4: [00:14:20] Slow to get off mute there. So this is.

Harpreet: [00:14:23] A real pain point question for me.

Speaker4: [00:14:26] For two reasons.

Harpreet: [00:14:29] I'm a freaking idiot is.

Speaker4: [00:14:30] The first reason.

Harpreet: [00:14:32] The second reason is, Hey, Avery, you didn't.

Speaker4: [00:14:36] Need to agree.

Speaker3: [00:14:37] So strongly on.

Speaker4: [00:14:38] That, dude.

Harpreet: [00:14:39] No, but the second one is it's just poor collaboration. Now, let me let me address both of them first. If you're an expert entrepreneur, you will never make the mistake of developing something the market doesn't want. Yeah. But because we love cool.

Speaker4: [00:15:01] Things and we've got great ideas, at least.

Harpreet: [00:15:04] According to ourselves.

Speaker4: [00:15:06] We'll go off and build.

Harpreet: [00:15:07] Something that not only does, no one know how.

Speaker4: [00:15:10] To use. But they don't.

Harpreet: [00:15:13] Even want to use it. And I learned the hard way. Okay.

Speaker4: [00:15:18] Why don't you go talk.

Harpreet: [00:15:19] To stakeholders and get a feel for what's perceived to be most needed? And then build that and it might actually help your ratings in your job.

Speaker4: [00:15:30] And make you a hero at your business.

Harpreet: [00:15:33] So I had to.

Speaker4: [00:15:35] Learn all that the hard way.

Harpreet: [00:15:37] But even if you are working on something that you think's not glorious and you do a good job at it, you will become a hero. If you're good at frequently checking in, hey, I'm not all the way there to what you asked me to do, but this is where I'm at. Do you feel like it's on track? And I have been shocked every time where people go, This is amazing. [00:16:00] You've gotten that far already. In a way. This is nothing. It's amazing. I don't know if you guys are this way, but I'll put my bar here and people only wanted me to reach here.

Speaker4: [00:16:13] And they would have thought that.

Harpreet: [00:16:14] Was a home run. And I've had to learn that so much. But the other thing. Is if you're doing good collaboration like that with the business to find out what they think's needed data wise. You'll find you don't always need a model. Just a good automated report on something, not even a full dashboard. You're just giving them frequent.

Speaker4: [00:16:38] Updates.

Harpreet: [00:16:39] On data. And then I've also just seen you don't always need machine learning. Let me.

Speaker4: [00:16:47] Explain this.

Harpreet: [00:16:49] We go through a lot of data processing just to get to machine learning. But along the way, if you're doing it very pristinely, you are going to.

Speaker4: [00:17:00] Reduce.

Harpreet: [00:17:00] It down to essential features and you're going to scale those features, get them on the same level playing field, and then you're going to be able to present whether.

Speaker4: [00:17:11] You.

Harpreet: [00:17:11] Use a simple model or not. This is the parade of feature importance. Now, if you think about modeling in general, the organization appreciating the parade of importance of features is much more valuable than telling them You're about to hit a pothole. Oh, wait, you're saying we could go.

Speaker4: [00:17:35] Down a whole new road and avoid.

Harpreet: [00:17:36] All the potholes just by understanding these feature creative influences? Yeah. So. In other words, all I'm saying is don't rush to giving predictions to the business. Tell them I'm doing this pipeline. I'm doing this data processing.

Speaker4: [00:17:55] Here's a parade of feature.

Harpreet: [00:17:57] Importance.

Speaker4: [00:17:57] Performances. I would think that's.

Harpreet: [00:17:58] More valuable.

Speaker4: [00:17:59] To.

Harpreet: [00:17:59] You than [00:18:00] the predictions, but I'm going to give you the predictions, too.

Speaker4: [00:18:03] But then I agree with the other.

Harpreet: [00:18:05] Things that are said. Infrastructure is everything.

Speaker4: [00:18:08] So if you can.

Harpreet: [00:18:10] Get it in production, well, that helps. But. Don't aim to develop something for production unless, you know, people have asked for it and are.

Speaker4: [00:18:21] Anticipating.

Harpreet: [00:18:22] It and are wanting to use it. Once you give it to them. We talk about trade or just like the 20% of whatever that gives 80% of themselves. Thanks for clarifying. So you're going through. Develop your your. You're going through the machine learning pipeline, which means.

Speaker4: [00:18:44] You've got.

Harpreet: [00:18:45] Your data cleansing routines automated. You are making.

Speaker4: [00:18:50] Sure you've looked at your distributions.

Harpreet: [00:18:53] For each feature and you've applied appropriately appropriate scaling routines to all of them so that.

Speaker4: [00:18:59] They're all in the same.

Harpreet: [00:19:00] Numerical level. At a certain point, you're going to find out which features are linear. You're going to keep the of the linear groups, you're going to keep the most important feature. Now you're going to have your base set of features that are most important, and you're going to be able to use feature engineering on those, etc. But at a certain point in that process, you will get to see this feature influences this prediction the most and this feature the second most in this feature the third most. And just from my business activities, if I had a data science come and tell me, Hey, I've got these predictions, but I've also got the relative importance of the features in the model. Let's say I want to see the relative importance of the features more than I want to see the predictions, because I can go work to act on that understanding of those features. [00:20:00] It's different. It is, I say, the term credo. All I mean is.

Speaker4: [00:20:06] This feature.

Harpreet: [00:20:07] Is most important, this feature. Second most important.

Speaker4: [00:20:10] This feature is.

Harpreet: [00:20:11] Third, most important.

Speaker4: [00:20:13] To this.

Harpreet: [00:20:13] Variable you're trying to predict. Awesome. Thank you so much, Tom. Appreciate that. Yes. I would like to add to this. Please do. You are currently on mute.

Speaker2: [00:20:27] So how about now? Good. So one thing I wanted to add, which is really a very critical piece, is. You know. One thing that we want to be doing any time you develop the model we are talking with predicts and any type of analytical solution is the end of the day is all about decision making, right? So I think what I have found is is a very fascinated users if they can get. Inside that can make them a decision making process. All of a sudden, adoption basically goes goes higher in those situations as well. So I think this would be part of the solution, exactly how the user is going to use that information and that can help them in the decision making process no matter where they are part of the process. So I think that can add the element is, is that if it is helping them in any way decision making process, the likelihood of that being adopted is much, much higher.

Harpreet: [00:21:31] Thank you very much. Great. We appreciate the insights here. For those of you just joining in, just tuning in on YouTube, on YouTube, on Twitch, wherever it is, was not. First few minutes of this conversation you were just talking about why? What's so special? What's so unique? What are some non-obvious reasons why only 20% of these projects seem to succeed? So if you're interested in hearing more, do not worry. This podcast episode is recorded from a YouTube channel, [00:22:00] so go and subscribe to the arts videos on YouTube. It's also going to be published on the podcast in just a couple of days, so keep an eye out for that. All right, cool. So let's go and hear from within the system then. Let's hear from you about the coming recession that you had, this post that came out of those talking about the bunch of layoffs that we've been having over the last few weeks. Like why all at once? Why is everything happening all of a sudden? Like, do these companies work together to just say, we're going to flood the talent pool? Talk to us about that. And then if anybody has any insight on this or anything to add to that, then take it away.

Speaker4: [00:22:48] It's been running. I hate to say this, but it's kind of been building up since last year. Around summer time, there were I think everybody figured out at the same time that they were their growth numbers were starting to decline. And it was really led by sort of the attention economy companies. So social media streaming, they were the ones who saw it coming. And a lot of it's a correction, it's a return to sort of pre-COVID numbers. Everybody went nuts during COVID. Anything digital, anything. Technology. We had a ton of digital transformation, spend get pulled into two years that should have been really spread out over five or six years. And so everybody's stock price went stupid. And I mean, legitimately, if you look at valuations, they were insanity. The numbers didn't make any sense. And so that started slowly decomposing and falling apart. And what's really been supporting a lot of the hiring sprees are the access to really, really cheap cash. And everyone's valuations were so high that it [00:24:00] was like they couldn't do anything wrong. The stock price kept going up. It didn't matter if they were their initiatives were returning a ton of money or if a lot of the speculative stuff just wasn't working out. And that's where we were at the end of last year. And then everything beginning of this year, just valuations came back to it's like everybody kind of realized, Oh yeah, like companies have to make money. Oh, and we've invested a ton of money in companies that don't have a path to profitability. Like Uber just came out and said, Yeah, we may never be profitable. That's just one of those.

Harpreet: [00:24:34] What.

Speaker4: [00:24:35] You know, if you put that in a in a quarterly report and just it's mind bending and that's really where we've come back down to is this reset to reality where profitability is now extremely important. So when you look at the dot com bust and it's funny, I I'm hearing a lot of people say dot com bust, but I don't think this is going to be that severe. I think this is just one of those speed bumps. Crypto's got a lot of troubles bringing. That's kind of what they call a contagion where a lot of tech companies have investments in crypto. And so in crypto starts unwinding, those investments go from breakeven, not really dragging the company down too much to all of a sudden. Now they are. And so they're looking longer term at getting return on investment from their crypto investments. Metaverse is slowing down and I mean, everybody was saying we're going to see a slowdown in the metaverse, there's going to be a bust. And then in two or three years, that's when it's going to be the massive opportunity. And, you know, and right now, the Amazons and Facebooks of the metaverse are being built, you know, right now. But there's still two, three, four years down the road. And it's like every investor just didn't listen.

Speaker4: [00:25:48] And so what's happening now is we're having this decompression where companies stock prices have been destroyed. And I mean, I made a joke about this on Twitter. You know, if you miss earnings, your stock price gets [00:26:00] pummeled. If you miss Earn or if you make earnings and issue lower guidance, your stock gets pummeled. If you make earnings, maintain guidance. Everybody that's been looking for a way to get out of your stock sells your stock and your stock gets pummeled. And it's just there's no win at this point. And that's what you know, it's these macro factors and these bigger economic factors. That is what all of these technology companies are reacting to. And there's a trickle down effect into more pragmatic businesses. You're more legacy business models as other competitors are slowing down on the really big, really ambitious data science and machine learning projects that are break even maybe two years, three years down the road as they begin to slow down investment because your investors want to see revenue and free cash flow now. So as that investment slows down, everyone is really doing the same thing and resetting and saying, okay, I need growth in free cash flow within the next six months or my stock price is going to get destroyed and the board is going to fire me.

Speaker4: [00:27:02] That is what every C-suite or right now is thinking is how much runway do I have with all of these investments that I put a ton of cash into before the board starts questioning my leadership and my inability to begin to improve margins and begin to show short term returns on data science, that's where we're in trouble. That is where you're going to see a whole lot of data science teams start taking a hit and start seeing downsizing and layoffs. I've had four people or four companies reach out to me about they call it rightsizing or it's really downsizing by a cooler name. I have had one of those in five years before that. So that gives you an idea. Of sentiment everyone is thinking about from a business standpoint if the business unit is not profitable. How do I begin to wind down pieces of it? How do I begin to reduce costs? [00:28:00] And we start out by I mean the good leaders are going to start reducing budget. Like we're going to close positions. We're going to reduce the amount of money that we spend on software and infrastructure. We're going to close down the.

Speaker3: [00:28:11] Travel.

Speaker4: [00:28:12] You know, no more expensive dinner know. But that's only going to last you for about three months. And then we're going to have to start laying people off and you're going to see it at every single tech company. And then there'll be a trickle down effect. And it's more pragmatic companies that really overspent and they have really profitable data science teams that they're not going to touch. Those are going to be those are the golden geese that have been performing for them. But they've been speculating and saying, you know, here's a three year better, here's a five year bet. And so those are the teams that are going to take a hit. Any team right now that is not profitable. You know, and I'm saying I'm telling a lot of people treat your team like a startup with three months of revenue left, you know, and that's really how you should be running your team is if you can't get to profitability in three months, there's going to be some pain.

Harpreet: [00:29:07] I have so many questions that are really stretching to take this. Let's. Let's just take it in this direction. So we're just now talking about teams getting in or getting cut off. What are the qualities of those teams that end up getting let go and cut off? Is it a lack of capabilities issue for the team? Does it have anything to do with the team at all? What's what are some qualities of let's say you are a company, an organization, you've got several data science teams. It's kind of make cuts there. Scientists are expensive. You need to go there and just do some expenses. Which one of my teams are correct?

Speaker4: [00:29:58] Right now, what companies [00:30:00] are leaning towards is just looking at profitability. How much revenue does this team generate? Because if you do cost saving initiatives, the problem is you don't need to maintain a cost savings initiative. Once it's deployed to production and it's stabilized, there's a level of reliability to it. You don't need the data science team to maintain it anymore. You can have a team of analysts, maybe even a couple of ML engineers who can maintain a large number of cost savings projects. And so it's not really capability. You can be amazing at delivering, you know, these these winners that preserve margins for the business. But if you don't have revenue attached to you or if you don't have a good long pipeline of cost saving projects that are you're working on, that will continue to return significant cost savings to the business. That's when you're in. The biggest trouble is because the C-suite, for the most part, doesn't truly understand exactly how you generate value, what opportunities they could be going after. And so there's a chasm of sort of value communication. Some data science teams and some leaders are really good at just hammering on, Here's the value that I create, here's the value that I create. They have good quality CDOs or CDOs or CDSs at the C-suite level who are again hammering home value, value, value, value.

Speaker4: [00:31:23] Those are the teams that are going to be much safer. And so it's a combination of capability, your roadmap and your leadership. So what you're looking for as far as a stable team is just that sort of trinity of things that you're working on, main business problems facing customers or whatever, the business model, those main line projects, you're supporting them, you're beginning to generate revenue and you're very much part of the conversation at the C-suite level, not director or VP. You have [00:32:00] to have someone who's communicating with the C-suite and who's effectively advocating for you, who's consistently hammering home the value of the team and not allowing somebody else from another team to kind of take over the narrative and say, you know, this budget would probably be better spent in my team. And that's what's happening in a few different companies right now is the CTO is going, you know, that budget I've been staring at for a while. I know how I'm going to get some of that. So you have to have an advocate to. Those are the three things I'd say I'd look for.

Harpreet: [00:32:35] Didn't think so much. If anybody has any follow up questions or comments or whatever. Then I say, go ahead, let me know. Comments on LinkedIn, on YouTube, Twitch, wherever it is you're watching here in the chat room. If you got anything to add, definitely feel free to raise your hand. And right now, that deejay with his hand up. Let's go to China. And then then I guess if nobody else has questions, the question I wanted to ask is and maybe we'll be able to get to it. But I guess what happens like innovation when companies start stop spending money like that, that's what does that mean? What does that mean for technological innovation? Let's go to Vijay, then we'll go to Jena. Avery said had a funny analogy that will go, go, Vijay, Jena and then Avery afterwards.

Speaker2: [00:33:30] Yeah. I was I was going to say, Ben, you are spot on. And I just wanted to. Kind of come back with additional data points. What I have been noticing, noticing. Last week I was talking to someone senior up in one of the startup company and they had to let go around 30% of the product data, product teams. And I think the justification that was given and some of the folks who [00:34:00] were kind of let go completely surprised, it just came. They just couldn't believe it happened. And one of the justification was that, you know, okay, who's adding to the revenue, right? Product teams. The product is there. Right. And you are not going to the next release. Let's survive with what we got. Let's keep the people who are kind of supporting the business and then just go and cut it. Right. So I think it comes down at this point of time, is that who can support the business going forward, you know, to be alive? Anybody else? Sorry. So I just want to make a point that that's exactly what is happening in the industry. At least two situation I have gotten to know.

Harpreet: [00:34:48] Thank you very much. Shout out to Joe. Joe something we have just joined as a top topic of discussion that we just not very nice this time around. These recent layoffs have been having happen at tech companies and what that means for data science and innovation kind of going forward and just what implications that has on our field. Gina, let's hear from you. And then if anybody has anything to add or riff off of, please do let me know in the comments. Raise your hand, drop something, LinkedIn, chat or YouTube and be happy to see it. Go for it. Cool. Thanks. Yeah. So Ben answered one of my questions in the chat, which was how are tools companies doing? Makes sense. Then you're saying they're doing pretty well. So when we say tools companies and you can probably articulate it better than me, but those kind of things that can sort of automate the data science and the pipeline building process. Would you characterize it that way? Yep. Okay, cool. So, yeah. And then. Then both your posting [00:36:00] in this discussion reminds me of a few things, and it just, it just tells you that we see the same cycles repeating in business often. So quite a ways. Back I worked in for a solar division of a big roofing company, and at that time solar was quite hot because California started the solar initiative, which had some, I think, very well thought out incentive structures and subsidies for solar to be more widely installed.

Harpreet: [00:36:32] Meanwhile, the backdrop in the industry was that there was a lot going on in Germany, Spain and elsewhere, and so prices were going up and up and up on cells and solar cells. There were shortages which drove things up further. And then another backdrop was just the economy was so hot at that time that things were going crazy. And then, of course, when the housing market fell out, a lot of this stuff just stopped. So companies went into that cost cutting mode. And for a lot of people who were in solar, especially if they were trying to do existing home retrofits, that was really tough because people were funding it with all their home equity surplus. And then when the market just went, then those businesses just mostly ceased to exist. But then guess what? After a while people kind of got things sorted out. They got there, they got more innovative financing models, much more like buying a car, let's say, than just then financing everything up front. And slowly but surely, that industry built itself back. So once again, it's kind of a thing of people rushing in. It's like a land rush or a gold rush, and then there's the bubble burst or those fallout, and then over time, that slow and gradual growth starts [00:38:00] to happen.

Harpreet: [00:38:00] And that's where the sustainable growth comes from. And then one other point quickly is that I worked for a great big company and I had shifted roles into a worldwide organization. And when new leadership came into the company to a point been made in his blog post, you know, watch out when new leadership comes in, especially if they're on a cost cutting thing. And Wall Street also tends to reward that, oh, there's change in this company's too bloated and we got to cut people. And especially in a big company, if you don't have advocates or if you're a cost center. And I was in a worldwide unit that was a cost center, even though we have long term programs that I think would be really successful. The return on investment was a few years out. So I mean, pretty much that well, the whole group pretty much got cut. So but I think the key points I wanted to make is just these are cycles that we see again and again and there's going to be fallout. So just a few anecdotes to underscore the point that it's where you're providing the value. That's the thing. Figure out how to do that no matter what. Thank you very much.

Harpreet: [00:39:13] Comment from one of my former colleagues at Pricing Matters. 140 solar was quite hot indeed. Thank you very much. Appreciate that. Let's go to ABC to get a funny analogy. By the way, if you have a question whether you're listening on YouTube or on LinkedIn or right here in the chat, whether it's related to or talking about mouth or something else, feel free to let me know. We'll get to your questions. Right now, we're just talking about the implications of this recent just this last week of all these tech layoffs, what that means for us as data scientists and maybe even what we could do to make sure that we are not one of those laid off. So [00:40:00] obviously kind of interesting times. Stock markets not looking great. Obviously, these layoffs not great. I'm not really old enough to remember 2008. Not old enough to remember the dotcom bubble. But it's interesting because before I started doing consulting, I worked as a data scientist for ExxonMobil, obviously not a tech company, but in 2020, you can imagine when everyone was going digital and working at home that was not good for oil and not good for ExxonMobil. So we had a lot of layoffs in 2020. And I kind of I guess I saw the opposite of like if like Facebook and tech companies where we're getting more traffic and driving in more revenue.

Harpreet: [00:40:44] No one was driving and no one was flying. So Exxon was making a lot less money. And so we had a lot of layoffs while there's a lot of hiring at other companies. So a lot of people left and got jobs at Amazon. So I think it all worked out for those people. But what's interesting now is in our stock price, pretty much halved. So most the time when I was at Exxon, the stock price was like around $70 and it got as low as $35 just because of COVID. And then also due to there was kind of a conflict between Russia and Saudi Arabia and anyways, that made gas prices and oil go down. And so a bunch of layoffs happened. Right. But what's interesting now is if you go look at the Exxon stock price, we're close to a five year high. And my my point just bringing this is it's kind of funny and I know oil is much more cyclical than tech, you know, but obviously this stuff happens. And like even though like Exxon was down two years ago and tech was up and now it's kind of flip flopped a little bit. And all of a sudden, like Exxon's hiring, I think these things sometimes like they just kind of happen, right? And I don't think at least I'm not personally too too worried about it because, you know, give it a year.

Harpreet: [00:41:54] Give it two years. Give it three years. I'm sure. I'm sure. Maybe, maybe, maybe. Vince, that is also [00:42:00] true that like we were already, this is just a correction down. It's not necessarily like a big swoop, but I just think in all businesses, there's there's good times and there's bad times. And I don't I don't think, at least for an industry wide, there's too much to worry about. Maybe more specifically, like, for instance, I don't really believe in Facebook too much meta or whatever that may be. They're, they're, they're hiring freezes might be a little bit more worrisome. But I think as an industry, there's the ups and the downs. And my point is when when one when one year you're down, the next year you're up. So I'm not too worried about a person. Maybe thinking that much. You might not be old enough to remember 2000, eight or 2000. You are definitely old enough to make me feel old. So just for context, you talk about Exxon stock. I was just looking it up right now, March 20th, 2020. So just a little over two years ago, it was at 32 bucks and today it's hot, like almost 89 bucks. That's why I was crazy.

Speaker3: [00:42:59] Well, it was also a negative back then.

Harpreet: [00:43:01] Yeah. Yeah. Let's hear from you. Let's go for it.

Speaker3: [00:43:06] Yeah. I think the thing I'd be paying attention to is just the rates, interest rates, that's what kind of drives everything. A really good book you should read actually is Econ How America? I don't know. It's basically I can't remember the subtitle of it, but it was written back in 2002. And it's about the dotcom crash. It is. And I read it a few weeks ago. And I think it's eerie how many similarities there are in the dotcom bubble to today. And I won't get into specifics. I think I'll talk about what's changed and what's the same. But first of all, what's the same? The tone of speculation on companies that basically made very little revenue, or if they did, then they made very little of these other weird things called profits. But because there was basically excessive money in [00:44:00] the system. And so it was easy to get money, easy for investors to play capital. And the other thing that was really big at the time was online day trading was had become a thing that had never happened before. And so for the first time, you had the middle class had access to spending. An enormous amount of money. Yes, that's a book there. Yeah. Actually posted in the shot. So, you know, and so this confluence of stuff led to a huge amount of capital to fund pretty crazy ideas, I think. What's then? That party was brought to a halt by one excessive speculation, which always ends badly in the end.

Speaker3: [00:44:39] And then the Fed raising rates because there was surprise, surprise hints of inflation going on. Right. And mismatch of supply and demand of specifically labor. And so fast forward to the financial crisis, very much a similar thing, low rates, easy money chasing what again, this time houses. You know, it wasn't tech, but it was the same sort of speculative mania. Anyone who was around at that time, it was just it was a crazy time. I mean, you saw your neighbors getting rich just like you did the dot com days. And the thing that always I always look at is when you when you feel like your neighbor is a moron, but your neighbor is making millions of dollars off of something, then you're like, am I the moron? And so that ended badly. But what happened was the Fed lowered rates to near zero, you know, and that's and I say for the last 13 years that spawned what you what you've seen out was just a period of insanely low rates, too much money chasing few things. And what do you think happens? This is what happens then with COVID. What I mean, like I think we're 40 or 50% of the money supply was printed in about a matter of a few weeks. So think about that. I think Warren Buffett calculated, you know, when I was at the Berkshire meeting a couple of weeks ago, he said these are $7,000 for every man, woman and child in America was printed.

Speaker3: [00:45:58] So think about what that does. So in [00:46:00] the last couple of years with COVID, you've had an insane amount of money chasing too few things. That's why you have inflation. And so what that means, well, what are you going to do about that? You've got to raise rates. So that's the thing I'd be looking at now, of course, are the rates. It's very fascinating because I believe Chamath Palihapitiya and the All in podcast, he had a good figure where it was like for every 1% increase in the Fed's rate, you can expect that basically a 15 to 20% markdown in valuations of startups. So if you get to a Fed funds rate of maybe, say, 3%, you know, you're looking at a pretty nice haircut coming from zero. So it's 45 to 60% off. So think about what that does. So what I think is going to happen is going to see a ton of down rounds and anyone who's lucky enough to get those, the rest are going to be either a purchased or. I'll probably die off. So that's, I think, what's going to be happening. So rates Warren Buffett always said, you know rates are the Warren Buffett. Yeah it's you know rates are rates acted as gravity on asset prices it's as simple as that. So that's what's happening now and it's going to be pretty bad.

Harpreet: [00:47:07] So.

Speaker3: [00:47:08] But I think good things come out of this because you can't you just can't keep having this excessive speculative mania. It's insane. So.

Harpreet: [00:47:15] Anyway. Thank you very much. No, no, don't so much talking about making money. So if you grew up, I believe them because they were in fact, let's go to them then. Chomping at the bit to add to this, please do let me know. I'd love to hear your perspective. Let's go to A.B. thanks so much for hanging out.

Speaker4: [00:47:46] I just wanted real quick piggyback on what Joe said about the rates. The scary thing and the reason why rates may go significantly higher than anyone expects is because right now the drivers of inflation. Fed doesn't control them. That's the real scary [00:48:00] thing is that we've got supply chain issues and we don't know how long it's going to take to unwind those. We've got these cartels that are and I mean drug cartels. I mean, like, you know, there's not whoops, I shouldn't have used that word. But there are noncompetitive forces and noncompetitive markets right now that are driving all sorts of different things, not just gas and energy, but a lot of technology segments are noncompetitive. A lot of your staples are noncompetitive segments. You look at cable TV, you look at telecom, there are so many noncompetitive segments and it's going to take a long time because they're just jacking up prices not for true inflationary supply and demand reasons, but because they can you know, there's there is a definite I don't want to use crazy economic terms, but there's, you know, demand is pretty stable for things like a cell phone or for a cable-TV package or for a business to be able to get access to the Internet and so on. And so demand is will pay whatever it's. And so that's driving a lot of inflation. And the Fed is under pressure by a whole bunch of people that don't seem to understand economics, that are telling them, you know, raise the rates faster, faster, faster.

Speaker4: [00:49:19] And if they do, then the economy craters like we have a recession. And right now companies are preparing for a 12 to 18 month recession and not a huge one. They're thinking contraction is going to be small, mostly flat growth. And so if that if the Fed overreacts and listens to the opinions more than the data, things could get a lot worse. And so the conditions that Joe is calling out where every time inflation or interest rates go up, you take a hit from a startup's perspective. It's really any tech company that takes a hit. And so you're going to see potentially more aggressive [00:50:00] layoffs, accelerated reductions of spending, and those are going to hit data science teams really hard because if they're not profitable. The teams can get cut back to zero because if they're not making any money, you could lay the entire team off. And we've got a ton of people coming into the field at the same time as we probably are going to be letting go. A lot of people who have never delivered to production. When I was talking earlier, about 60 to 70 to 80% of data science teams have never put anything to production. We've got this really ugly potential perfect storm coming where our field, as secure as it is today, in a year and a half from now, we could be in a totally different place.

Harpreet: [00:50:49] For a year and a half, things might change. That is the future. So what do we need to do to future? Futureproof ourselves against a threat from Schumpeter on this one. Great shout out to you. Looks like you're driving everything. I don't want to bother you. I will keep you safe. But by no means you've got anything to add. Please feel free to hop off on Newton and. But make driving safer and more secure.

Speaker3: [00:51:20] Hi, everyone. Yeah, I just want to I don't have comments on the stock market or the economy, but the reality is that layoffs happen sometimes in droves and sometimes sort of more randomly. My general sort of feeling around this is that you have to be. Prepared, I guess, is a way, no matter how valuable you think you are to a company. To the company. I just got a new puppy. She came running to me. I'll show you guys real quick. Anyways.

Harpreet: [00:51:51] Cute.

Speaker3: [00:51:53] Thank you. Yeah. So the best thing that you can do is always be networking, always be making [00:52:00] important connections with folks in your industry. And, you know, just just be ready to have a plan B, be ready to activate those those networks and, you know, be ready to make a career switch in a short amount of time. So just never be too comfortable. I guess it's my advice.

Harpreet: [00:52:24] Shot that. Thank you very much. Russell, thanks for joining us. Appreciate you being there. Greg, anything to add at the top of all these layoffs is happening in tech. Can. Can you guys hear me? Yep. Yep. Yeah. So it's, uh, I was more on the listening end of this, and I think Avery made a point in terms of like, uh, how bad is it going to be? Uh, in my opinion, is it going to be another 2008 where.

Speaker4: [00:52:57] People.

Harpreet: [00:52:58] Like, like, like, really did something bad that affected, like, I consider the whole world. Um, I, I'm not too sure. Um, but at the end of the day, this is a sign of, you know, re re, I guess, readjustment and is what, in my opinion, it should have been. Expected because we were doing so well or a lot of companies were doing so well when the when COVID hit. Right. The first announcement to me was, you know, Zillow, then Peloton. Right. I don't remember like who did something about Peloton, but I was like, how can you be a CEO and know that you have an anomaly? You you consciously chose to go after that anomaly to prioritize revenue and hence profits without knowing [00:54:00] that once we were actively trying to solve this pandemic, things will get back to normal, thinking that you were going to remain an anomaly forever. So that was quite, you know, like I don't understand how our greed can just overshadow things that.

Speaker4: [00:54:18] Could have been avoided.

Harpreet: [00:54:19] With a simple, you know, data analytics or, you know, so, so to Joe's point, you know, it makes you think that who's really the driver here? Like, is it really the Fed's fixing the interest rate or something beyond that? Right. Can we call it greed? I don't know. I can't put my finger on it. But there are some forces that, you know, make us take decisions.

Speaker4: [00:54:44] That.

Harpreet: [00:54:45] We know can be fixed over the long term. Right? Oh, if the market crashes, somebody will profit either anyway and it will readjust itself. Right? So those are the things that we.

Speaker4: [00:54:57] Should.

Harpreet: [00:54:58] Definitely keep an eye out on.

Speaker4: [00:55:00] For example, on the personal.

Harpreet: [00:55:03] Side, I invest whether the market goes up or down, I invest, right? So I.

Speaker4: [00:55:08] Put a couple of money, a couple of dollars.

Harpreet: [00:55:12] Here and there. And what that gives me is, I guess, peace of mind that this is a cycle, right? In these cycles when things are bad, you see those layoffs which are super unfortunate.

Speaker4: [00:55:24] And to I.

Harpreet: [00:55:26] Think, Vijay, your point, too, is that who needs to go? Who's the most important person to stay? How do companies decide to do that? In my opinion, it's really culture, culture driven. Um, in some companies they say without a sales team, you can't make money. And some companies without the innovation team, you can't do anything, and some other companies without the other team, you're not you can't generate anything. So it's really culturally driven. But I've seen like. Seems [00:56:00] that have the projects.

Speaker4: [00:56:03] That could have been are.

Harpreet: [00:56:05] Typically affected first. Right. So if it's like an idea that you had that could have like worked out, they go straight to it first and eliminate that and say, okay, we can take a pause because our competitors are going through the same thing. So let's delay that to next year. Right. And that's what drives them. Let's say a competitor didn't.

Speaker4: [00:56:27] Really eliminate any workforce.

Harpreet: [00:56:29] They're going straight on that innovative research or that stuff. They would have probably swallowed that pill and go next to the next pocket. Right. The next pocket could be some cost top project they may be having. Or we rehired too many people in the upstream. How can we go lean there. Right. And then they go there. Right. So it's kind of like everything. It's a combination of emotion, which is I'm adding greed to it. And then a straight, you know, I would say strategic business kind of thing. But emotion is the leader there in that in pregnant.

Speaker4: [00:57:00] Into, you know, the.

Harpreet: [00:57:02] Company culture. Hopefully that made sense. Sorry, I said too much. No, no. I love it that you think it off at that point about the innovation team. I mean, I wonder if that's like a place where companies would consider not cutting off because we need the innovation team to figure out how to do the work of all these people we just laid off, but yet do it for free or reduced costs somehow. But anyways, let's go to Jo. Jo, let's hear from you. By the way, if you're watching on LinkedIn, on YouTube, on Twitch, if you're here in the chat and you got a question, please do let me know. I'm happy to take your question.

Speaker3: [00:57:42] Well, I mean, just following on Greg's point, I think he's absolutely right. It's there's a lot of drivers and a lot of them don't really ever make any sense, but I guess they always do in hindsight. Right. And that's why in the league I posted a zoom here, a link to a book called Manias, Panics and Crashes A History of Financial Crises. [00:58:00] And What I think it behooves everyone to actually study this kind of stuff, because what you realize is there's nothing really new that's going on. More like we just keep making this dumb, dumb mistake, same mistakes over and over again and, you know, different versions of it. You'll never be exactly the same mistake, but you'll the same behavior and the same sort of emotional frenzy will get you into a different type of disaster down the road. And so it's really fascinating to to watch. I think, you know, we saw this recently in the whole, you know, in this recent tech boom, you know, startups and everyone, you know, how is it that based upon zero revenue, for example, or very little revenue, you know, you now have thousands of unicorns, companies that are worth over $1,000,000,000. At a fundamental level that makes zero sense whatsoever, right? Like if you use a traditional financial valuation models, it makes no sense at all. I say I saw valuations being done on the number of GitHub stars and the number of people in people's Slack groups. And I know people that have raised at valuations between close to $200 million and north of $1.5 billion based upon no sales and slack member accounts. So we'll see about that. But let that sink in as a data point and then. And then you understand, then then ask yourself, okay, so if these companies or maybe A, series B, series C, how are they going to raise money and what's the future now that valuations are most certainly getting marked down?

Harpreet: [00:59:40] So.

Speaker3: [00:59:42] What's it?

Harpreet: [00:59:44] It's like the companies getting valuations based on slack number user accounts. That's the argument that some websites had back in 2000 years ago. Exactly the [01:00:00] same thing. Joe, thanks so much. I appreciate that. Anybody have anything to add to add here? Feel free to let me know. Otherwise, I just had a question I was planning out with Snowflake today for the first time ever. How a world tutorial packaging integration that's kicking off with Snowflake. We're launching that in just a few weeks and. What's the hype about Snowflake? Maybe I'm just not that prolific of a sequel user to get it. What is it about Snowflake that people just love? Would love to love to hear from you guys. Anything to add here about Snowflake if not from Joe or something? Or Ben or Tom, please let me know.

Speaker2: [01:00:48] Yeah I was going to so earlier point I think Joe was making. You know, one thing that we we have seen this movie before and feel like we never learned from it. Right. And I think that one of the reason I can see is the emotions, right? When emotions are in the play, all your experience goes out of the window. And if that's what is happening. Frankly speaking, in this one and I, I just I was thinking my mind, how come we're still doing the same thing? We are in the same place. You burning off your money in the stock market, you're losing money. How do you not learn it? Right. So I think the most sense is definitely taking all the rational decision, how we behave and how we act, even at the company level as well to do that. So I just wanted to add that.

Harpreet: [01:01:35] Thank you. Thank you very much. Appreciate all that. And you know what? There might be risks that these companies take, but without these risks, we can't really drive innovation forward. We've got to let's make all of these kind of dumb decisions because that's how we keep pushing ourselves for. But Tom, let's hear from you. And after this conversation, I really [01:02:00] just want to know you guys thoughts on snowflakes. If you want to say something about Snowflake right there in the chat. Wherever you are, Tom. You know, Gina asked an interesting question, and I notice that I think it was cost. And we're we're.

Speaker4: [01:02:16] Discussing something.

Harpreet: [01:02:17] In the chat, but I'm really I have a thought that I'm interested in other thoughts. So as soon as you talk recession, you're worried about supporting your family.

Speaker4: [01:02:28] And it occurred to me.

Harpreet: [01:02:31] You know, Harpreet you and I were recently.

Speaker4: [01:02:34] Looking for new.

Harpreet: [01:02:34] Roles, and we landed in places we loved, but really got me thinking about some of the questions in the chat. Where do you go? Job market wise as a data scientist if you want to stay recession proof.

Speaker4: [01:02:52] And it occurred to me.

Harpreet: [01:02:54] Wow. Really? Like where I met. What we're working on. But. It. It's not going to suffer.

Speaker4: [01:03:00] If the.

Harpreet: [01:03:01] Economy goes down. It's a.

Speaker4: [01:03:05] Forget this tangent.

Harpreet: [01:03:06] But one night recently talking to the kids at dinner about why hey is important. And it got them thinking about Hey.

Speaker4: [01:03:17] Unintended.

Harpreet: [01:03:18] No. Hey, no pizza.

Speaker4: [01:03:21] Please make the connection. So it occurs to me there.

Harpreet: [01:03:25] Are some domains and I think Gina was bringing up the point in the chat. What do you think about these? Ml or.

Speaker4: [01:03:33] Data science tool companies.

Harpreet: [01:03:35] Know arrows aimed at where you're at harpreet. But. I guess it depends on what they're doing, how entrenched they are. I don't think data robots going away, for example. It just occurred to me if it's a domain that's.

Speaker4: [01:03:52] Not going to.

Harpreet: [01:03:53] Suffer because of the recession, you're probably in good shape. But. Yeah. Which [01:04:00] areas do we work in that tend to be less safe? I guess I would think a tools company that's very new may be really scared. And someone.

Speaker4: [01:04:13] That's serving an industry.

Harpreet: [01:04:15] That it just is absolutely needed. No matter the state of the economy, you're.

Speaker4: [01:04:19] Probably pretty safe, but that's pretty.

Harpreet: [01:04:22] Simplified model. I'm very eager to hear the thoughts of others on that one. Actually at some point time, I think. It's going to be jobs that require more creativity. Things that cannot be automated away. For example, developer relations, for example, develop. That is a job that yeah, it's going to be affected by the tech industry and innovation being pushed forward. But these are roles that are highly creative, yet they tend to see cost synergies. But the work that developer advocate relations professional does can't be like automated. I can't super it's super specific knowledge that is such a unique combination of skill sets that you can go to school to become this type of person, and yet this type of person is highly able to contribute a ton of value with not a significant amount of effort. But a lot of creativity. Let's hear from you if anybody else has anything to add.

Speaker2: [01:05:36] Yeah. So I was going to say so I'm in pharmaceutical industry and that's where my expertize comes in. What I have found is in pharma, you know, economy goes down, up, right? People are we are still not able to produce the medicine that we are basically want to give to the to the patients out there. There's unmet need. So [01:06:00] in a relative sense, I have seen that life sciences companies do well even there isn't time unless unless the pipeline is tried and sales is going down, there are certain of that nature as well. But I can relate to one industry which is pharmaceutical mostly for majority of the company that think they don't get impacted that much by this kind of changes. So that one thing comes to my mind.

Harpreet: [01:06:33] Thank you very much, Vince. Let's hear from you.

Speaker4: [01:06:37] As far as talking about what what industries are safe? What companies are safe? I think in the near term, just look at the cash position. You know, it's not even sector. It's not even really segment reason why a whole lot of startups right now are just dropping as many people as they can and getting as lean as they can is because they're assuming that they can't raise any more cash for the next probably 2 to 3 years. And so they're looking at how am I going to either reduce my costs or get a little bit more margin out? You know, can I raise prices? Do I have any pricing power? Can I they're trying to survive three years. And so any company that's in a good cash position to make it through with stagnant growth or maybe a small recession like 2%, 2% loss, 3% loss or contraction or whatever. Yeah. You know, anybody that's got that kind of cash to survive that is probably in an okay shape. You're going to see layoffs no matter what. I mean, shareholders are going to almost, almost force it because what you're seeing is institutions cycle out short term investments investors are cycling in. They're the ones who always tell you to do the short term things like lay people off who you really shouldn't because you're going to need to just hire them in two years. So it would be cheaper just to keep them. [01:08:00] But so you're going to see a little bit of short term thinking at every company. But every company that has a cash position right now that'll get them through the next two or three years is a good you know, it's a good bet.

Speaker4: [01:08:14] The thing that I'm seeing probably about three years down the road is kind of what you're saying is they're going to stop investing in innovation. And the companies that stop investing in innovation are going to get leapfrogged by companies that continue to do it. Undercovers. And what you'll see is you can't talk to investors about the amount of money that you're actually spending on innovation. And so when we go into these sort of recessionary periods, some companies are smart and they keep the innovation pipeline going and they keep moving forward with these projects. They just slow it down a little bit and they'll always couple it to short term gains, short term wins. They'll figure out how to spin off pieces of it that make profitability, you know, make it look like a normal business unit, even though under the covers there's a lot of innovation going on and those are the ones that come flying out of the gate. And what I think you're going to see is in three years, when those companies come out and they start really throwing out new and innovative product lines because growth is back and it's okay to talk about long term bets, other companies are going to be so far behind because they couldn't do that, that there's just going to be a die off. And I think what you saw happened to retail right after we came out of that first recession is what's going to happen just across industries. I think you're going to see like a retail die off, but everywhere. So short term safety is cash. Long term safety is innovation and responsible innovation.

Harpreet: [01:09:47] I didn't think so much. Chaplin said in the chat that the tech industry specifically is very skeptical. Two weeks ago, it was impossible to hire. Now there are layoffs, so there's [01:10:00] a lot of action going on in the chat. If you guys want to be here in the room, you can check out HTTP or bit.ly. That's a bit early for us. Absolutely. Love to have you guys here in the half hour live and direct with us. Also, again, the huge financial sponsors and the World Machinery Production Conference happening June nine, June ten, 1130 in Toronto. There's a link to register for the event. The comments use the promo code HARPREET or the capital H. 50% off. Let's go to coast to coast. Good to hear from you, man. Good to see you here. And then after that, we will go to Egypt. Yeah, I guess kind of on that. So I was, I was thinking a bit of a long, hard look at where my career is at and like this is maybe a year ago, right? Just over a year ago where I started realizing that the robotics industry is kind of weird, right? The robotics industry is a high upfront capital investment. It just needs it, right? The sheer cost of manufacturing, of even prototyping, the robots, the electronics that goes into it, very high capital investment upfront. And typically you don't see that payoff until a number of years down the line because you need to reach a certain amount of scalability. Now I'm talking more field robotics than, say, process automation or say industrial automation, but that's where I kind of looked at the field robotics industry and I went, Okay, how do I, as a computer vision engineer, see that? Like see this through where I'm going through this almost a lean patch economically where some small robotics teams are doing really well because they have a very specific job within, say, oil and gas, for example, where they're doing, you know, I can name one or two here in Sydney and they're doing really well.

Harpreet: [01:11:54] But the other robotics companies are struggling to find capital. They're struggling to, you know, and we started seeing this [01:12:00] maybe just over a year ago in robotics. And now I think it's starting to hit the wider market. And like, this is the first time. I mean, I've only been in the workforce since like 20 1617, right? So I'm really new to all of this and I'm looking at this going, okay, so how does someone like me approach this behaviorally, right? What are the we've seen? So from a business perspective, like Ben was saying, if you focus on cash flow positive and you focus on essentially how do you keep as much of your R&D development for your future afloat, those are the key winning traits of the business. What are the key? Winning key. Losing traits of of an individual like a junior engineer or a intermediate engineer. Yeah. That's behaviorally something that I don't know, I've never experienced. I'm really curious what the room thinks. So the question there is I think the core question is what are the qualities of the individual, not the team or the quality of the individual that make them valuable? Like this really hung on. I think you had your hand up for actually.

Speaker2: [01:13:23] No, I had the previous comment I had made already, so I addressed that, actually.

Harpreet: [01:13:29] Yeah. Let's get back to me on this topic. We were talking about how to say future proof. So it's kind of doubling down on this question there about just just. Quality not only futureproof, but just invaluable. So you cut.

Speaker3: [01:13:52] Out a little bit for me. What was the second part?

Harpreet: [01:13:55] Yeah. So the the question that you're asking is, what are the qualities [01:14:00] of the individual contributor that will make them invaluable to an organization? Right. So I get that essentially. But I mean, maybe the easier question to answer is what are those behaviors or qualities of of an independent contributor that don't make them valuable or lose value quickly? Right. That might be easier to answer than say, hey, this is what makes people super valuable, right? Because that can come in so many different shapes for different businesses.

Speaker3: [01:14:32] It's kind of the cardinal sins, really, like lethargy, sloth. I mean, think of the qualities you like and you don't like in people, right? And then translate that to somebody you're paying good money to. I mean, I would evaluate it to that lens. If I if I didn't like you before, I probably ain't going to like it now, for example. So, you know, and if you weren't, you know, and I'd say that, you know, to answer your original question, if you if you look at what the qualities of what makes somebody, I would say, indispensable, it's going above and beyond and trying to and focusing on keeping the business going and keeping the business growing, preferably. I mean, it can happen in recessions. I've seen it myself, but it really it takes a different quality where you're going above and beyond and trying to say, okay, well maybe that's not in my department per se, but I'm definitely willing to pitch in and help. I'll keep that person around as long as possible. Right. It's the people who are basically like, you know, phoning it in, not making an extra effort. I would say, like if you think, you know, here's the deal. Just because you got a job that with a certain job description and when times are tough, you're expected to do more. That's how it is. And if you can't do more. Then I guess I'll either need to find somebody who can do more or I'll start doing more with less. So, you know, it's just that's how I see it and that's how I've operated in recessions in the past. So it very much is the eye is on like, what are you doing? What are you doing to help out? So [01:16:00] I would just say, like, make sure you just bust your ass showing what your capabilities and making sure you just are indispensable. That alone is going to get you that gets you far and great economies that get you even further when times are tough. So.

Harpreet: [01:16:14] So much shopping at the story. If anybody else has anything to add. Please let me know if you hear the chat question on LinkedIn or on YouTube or on Twitch. And if you have questions, you know, to drop them in the comments and will gladly just shove it. Yeah.

Speaker3: [01:16:31] I'll push back on Joe's point just a little bit because it's, you know, it's. Of course, those those who will put in the extra effort are likely to get recognized more. But at the same time, if we're putting that pressure on everyone, I mean, people have different priorities. You know, people have different family needs and, you know, and things like that. And work life balance is extremely important as well. So I, I just don't think it's super healthy to just like, put, put all you're all in to towards the goal of of your employer. And then, you know, even even if you survive, you know, the set of layoffs, I don't I don't think that's super worth it to like sacrifice other parts of your life to do that. What I will agree what I do agree vehemently on is if you mean by above and beyond, like making the impact of the work that you're doing obvious and sort of connecting the dots between different different departments, different needs, and really rolling up what you're doing and to what the business is interested in that I fully agree with. I just think that that can be done strategically rather than sort of just working away at it. Oh, yeah. I mean, as an employer, it'd be smart about how your what you're demanding of people. But, [01:18:00] you know. Yeah, I mean, we're more in agreement. They are in disagreement for sure. It's. You know, because. Because the thing is, too, if you treat your employees like crap, the first thing you're going to do when the times get better is you're going to go somewhere else, right? So and you don't want that reputation.

Speaker3: [01:18:14] But at the same time it's definitely here's a deal more is going to be asked of you is as simple as that and like you know I think it's interesting because a recent generation, I would say of like tech companies as coddled people, I think to a to a very uncomfortable degree, to the point where it's like they don't understand what it's actually like when you have a downturn. Like I've been through many recessions at this point, I've seen it and it's like this is there's peace time in this war time. This is war time. And unfortunately, when it's war time, you know, bad things happen. So I think the the days of like, you know, the things that irritated me back in the last ten years, it's been, you know, startups raising a bunch of money and, you know, ping pong tables and all this cute stuff, kombucha on tap and like massage therapists coming to the office and all this all this cute shit, right? I'm old school. I don't care about that stuff, you know, as a boss. And like, here's expectations, you know, let's make a fair agreement on what's expected of you. That's it. I'm not here to give you a freebie or I got Fridays or something like that. It's not my style. It's just like, you know, we have agreement. You here to work, you know, I'm here to help you grow your career. That's it. So, I mean, it's proven, too, that that stuff doesn't really help the employee either. That's all bullshit. It's really stupid. So, you know. But the thing I was joking actually with with Kenji about this, we're trading a schwag at my house like startup schwag.

Speaker3: [01:19:37] And it's like the really horrible thing is we're about to see the schwag pocalypse happen. Like all, all the startup schwag is going to disappear, and there's a lot of nice schwag out there, but it's like we're trading socks with each other and a bunch of other things and it's like, Yeah, I'm going to really miss these days. So, so anyway, now I think it's a great it's a great point you bring up Stoner I mean, it is. It is, I think because at the end of day it's easy. And I, I've worked [01:20:00] enough toxic environments where it was like, you know, in good times, your employer was just a total asshole to you, to be frank. Right? And just like they just abuse you in a bad time, it's like, yeah, what have you done for me today? Because if you ain't done nothing for today, then you can get out. So that's I've had that happen and I've left those places. I left I left a job in 2010 in like the depths of the the Great Recession, you know. And I had a kid on the way. I had no option. But I was like, I know I got enough money and I got enough skills. I don't need to put up with this crap at all. So, you know, I think it's an extreme case, but. You know, it was it was bad enough where I was like, so I've had that happen. I've had I've had losses so toxic. I'm like, I'm not dealing with you. Even though every rational person says I should deal with you, I'm not. So. Yeah, that's a great point.

Speaker2: [01:20:52] So, Joe, I would say that being asked more for someone, it's coming in a natural way in the sense that if you're letting some people go as a company, then you're asking someone to fill up the gap, right? So necessarily it is not going to be, hey, you're gone and that work is gone. There's some something's happening. So people will be asked to do more, actually. So it's a very natural kind of way of asking for due to fill up those some of the gaps.

Speaker3: [01:21:21] Exactly. Just make it worth the people's while, though. I mean, I have empathy with them too. They're going through. Yeah, indeed. They got it. Yeah. So it's like don't just say, oh, you had more work to do. Have a nice day. It's like, that's bad. So. I think that the pressure that comes from seeing your coworkers getting laid off and then using that as a motivator to work harder and pick up the slack, I think that doesn't beget psychological safety at work. I actually think that that's if that's a strategy, that's a bad one. It can be mishandled for sure. I've seen it. To the earlier question of what makes how can you make [01:22:00] yourself more replaceable, I guess. And from from what I see, it's much easier for senior engineers to hang on to their jobs than than junior engineers. So it's kind of an obvious one. Upskill and try to communicate the impact of your work, as I said before. And then if you even regardless of what level you're at in terms of seniority, like I said, don't put all your eggs in one basket. I think it is important if you're if you're replaceable, company should be replaceable to you as well. That's my message. I'm happy to talk about Snowflake when we get to that, by the way.

Harpreet: [01:22:48] Yeah, absolutely. I just need to take that one on one line. But thank you. I think, Joe, really appreciate that. Some of the points there. Just talk about how senior engineers are likely to be around longer than like junior engineers as a collective, almost like the effect type thing. Just because they've been in the game for this much longer, these are going to stay in the game for at least that much longer. But to a question like that, Richard, completely was how do you make sure? Or the question is, what do you have to do to make sure you get that type of thing? And I would say it's the tendency to be very, very quick to start something but incredibly slow to finish something, because it just leads you picking up things and failing business and productivity when really you're not really focused on anything. You just try to push up the uptake in August and to understand, but really completely nothing zero. I think that is definitely a surefire way to. She put [01:24:00] herself on the board. Let's go to Kosta and then Gina, and then we'll see what other questions people have and begin to wrap it up. By the way, if you have questions or comments, please do let me know right there in the chat after having heard from our wider tours that definitely come on. The funny thing is that most people today do 20 to 30 years old, do not understand or have extensive inflation rates and the impact of just free money for a long time. Tor also said that said it's hard. It is not hard to succeed or have to work hard harder than others towards the end.

Speaker3: [01:24:41] Can I just say a very funny tidbit about inflation for 1/2? So it's funny. I was hanging out with my my grandma the other day in Omaha and it was she wanted to be back in the back in the eighties, early eighties. Inflation was so bad that banks were giving you like a free shotgun to deposit your money in a bank. Think about that. It's crazy on a lot of levels. So, anyway, go on.

Harpreet: [01:25:11] I wanted to ask about their place because. Substitute engineer. I think Jen actually had her hand up before me. Yeah. Go for it. Right. Thanks. So, Joe. We were practically neighbors growing up. I grew up in South Dakota. Yeah, I have a friend in Omaha right now, so. Yeah, high school friend. So anyway. Yeah, you know, not to be. Maybe I'm just crusty and cynical and having a few years under my belt.

Speaker3: [01:25:50] But to echo.

Harpreet: [01:25:51] The comment that I guess to made, you know, the truth is, like all these companies can say all this nice stuff [01:26:00] about how they care about people and this and that and the other thing. But when the chips are down and when it's hitting the fan, yeah, you'll see. I totally agree with you. Sentiment, don't get me wrong, I totally agree with you. And what I love about the next generation coming up is like they seem to be quite serious about, look, you know what, I'm not going to give everything to a job. I'm not going to throw myself on to the fire for some job. I think that's a way healthier attitude than, let's say, some of the people, you know, and the thought process of my generation and the one before where it was like you were your job in many respects. At the same time, I just kind of want to inject yet another dose of realism, as Joe has, that it's going to.

Speaker3: [01:26:53] Happen and to.

Harpreet: [01:26:54] Some extent low end. And like you said, I'm sorry if I mispronounced your name, but it's natural that when when there are layoffs and stuff, that people are going to be asked to pick up more. And then you have a decision to make. Your decision is, am I willing to do this or do I have other options? And if you have other options or even if you don't like Joe said, I mean, I've been in toxic work situations and sometimes it's like, look, I don't care, I need to leave because I need to like I can't handle this anymore and I've put up with a lot of stuff. I'm pretty tough that way. But you get to a point where it's just like, This is BS and I'm not doing it anymore. But all I'm saying is, is like it's give and take. So we always have a choice, right? Even if our choice is to jump off the cliff into the unknown, we do have a choice as far as whether or not we're going to stick it out. And I think, as Joe was saying and been to this [01:28:00] is where you find out what kind of people you're really dealing with in the workplace. And so if you have no other options, you stick it out and then as you over say, the minute things get better out.

Harpreet: [01:28:13] So yeah, I just kind of want to add that. Oh, and one other point, which is sometimes even working hard, even sometimes that's not enough. It depends on the organization, how political it is. I mentioned being in a very large company and people were laid off wholesale because there isn't enough of a connection between what any individual does, the value they add, what potential they have versus we got to lay people off. And so the people who landed in other positions tended to know other people. They'd been in the organization for a while anyway. That's just what I want to. John. Thank you so much, Constable. Yes. I just kind of wanted to gather a couple of thoughts from what I've heard from across the room. Right. We've kind of I guess my generation, like I said, I've been in the workforce since 2016, 17. Right. My generation of employees, essentially, where we've kind of been very lucky that we haven't seen a recession yet, much in the way that, like the last couple of generations, hasn't seen an all out global war. Right. And in the last couple of decades like this is not we haven't seen a Vietnam War. We haven't seen a World War Two in a very long time, thankfully. Right. And I guess the optimism that's kind of built into us is while that's really powerful, can also be a weakness, right.

Harpreet: [01:29:43] If we don't back it up with resilience is essentially what I'm hearing, because like Joe was saying, you can make that choice to leave a job even in a recession if you know that you've got the frugality and resilience to deal with the resources that you do have at hand. And [01:30:00] making those choices is the difficult part. So how do you. So I guess the my big takeaway from this is it all comes down to resilience, right? It comes down to resilience and understanding what your values are and how you're able to trade off your personal values. Right. And what your plan is. And resilience is partially comes with like within some kind of end in sight. Right? So if you are trading off your values, how do you get back to acting to your core values? Right. So that's kind of what I'm taking away from this. And I mean echoes kind of what Gina said as well. So I think where we need to balance that, how do we how do we learn that resilience without losing that optimism? It's kind of going to be the key question for for me to take away from this. So that's going to be something I'm thinking about for the rest of the weekend.

Speaker3: [01:30:51] Thanks, guys. I mean, the advice I'd give people is, I mean, especially when you're young in your career, start building a big pile of, like, few money. You know, money goes a long ways to I think that's singlehandedly the biggest thing you should be doing, plus building your network. But it's like having money in the bank. Having that kind of emergency fund means that you have options. You have options to make a choice about how you want to spend your time and who you want to spend it with. If you don't have options, you know, things are just a lot harder. I mean, we're all privileged. We work in tech. We get paid well on a bad day. We'll probably find work, right? Like you got good networks, all this other stuff. Most people aren't like that, you know? I mean, go, go talk to people. I mean, I you know, I've run over in factories before, but people were like, they are making nothing, you know? And I was like, you know, I can't afford to pay more. I'm sorry. That is what it is. And that sucks. I mean, I do feel for that situation, but it is what it is. And like those people don't have options. And I've worked in, you know, doing manual labor and stuff [01:32:00] before and that's kind of like the career that some people are in and they are not getting out of that, you know, and I don't think they there's poor as a church mouse.

Speaker3: [01:32:07] That's kind of how it is like we got options. So it's like if and you have everyone here has a tremendous ability to, you know, Jesus hasn't yet amassed cash reserves and you know, even if however you do it. But that's my biggest advice is like, you know, build optionality and especially when you're young in your career and especially in the good times, it's tempting to go on to buy that Tesla. It's tempting to want to go and spend it all on, you know, fancy trips and all the other fun stuff. And you can go ahead and do that. I would say once you have your bases covered where it's like if something bad happens and it does, this should be a reminder things change on a dime. Right. As Santonio was pointing out in a chat a couple of weeks ago, everyone's like, It's so hard to find people. Well, it's going to get a lot easier. So. You know. But, you know, if you if you get your bases covered, if you've got a good network, if you've got money, you can do what you want.

Harpreet: [01:33:00] And capability at its finest. Thanks so much. Then let's see if you can get this one and if nobody else has questions or comments on that on anything after this. We're wrapping it up and I will contact you at one on one to talk a bit about Snowflake and also see what astronomer and package can be. Pretty awesome. Joe, thanks for hanging out here then. Let's hear from you.

Speaker4: [01:33:31] I think the greatest line I've ever heard from a movie is anybody tells you money doesn't buy, happiness, doesn't have any. It's the truth. You know, it's it's a lie. So, I mean, what's the best position you can put yourself in? It's the one where. Yeah. Boiler room. I had to edit out one of the words in that line, but yeah, it's, you know, if you put yourself in a position where you can start your own business, that's where you want to be because you [01:34:00] begin to every day generate wealth for yourself. And it's generational wealth. Having a business that you can pass on that has valuation and especially now when we see the M&A cycle that's going to happen, even companies like mine are getting offers like it's dumb what companies are trying to buy right now. They're trying to do aqua hires where they will buy a company just for its people because it has capable people and they haven't been able to hire the type of talent that they need and assemble the team. So let's just buy a company. So if you if you want to talk about who's going to be most successful at the end of in the middle of every recession, we have an uptick in business formation, business birth. This happened during COVID. If you look back to 2012, I mean, that's when I founded my business.

Speaker4: [01:34:54] If you look back to 2001, you know, and there are unicorns that keep showing up out of each one of these recessions. And, you know, some people are coming out of college that are starting. Businesses that survive do well. And not every business is going to be a unicorn. Many of the successful businesses that come out of each one of these are just businesses that do mid seven, low eight figures and they slowly over a decade grow into something that's, like I said, generational wealth. Now you have that money, you have that stability. Most of the the share of profit is not going to go to the employees. So when you talk about the value that's going to go to somebody, it's always going to the person who owns the business. And so when you talk about the best way that you can generate value for yourself no matter where you go career wise, it's always going to be creating your own business as the best possible way forward. [01:36:00] And so if you're an employee, you right now. That's what you want to start building his entrepreneurial capabilities. You want to start building the ability to innovate with value. And this is all stuff you can do inside of a business. It'll make you super valuable to the business because you'll be able to talk to the C-suite in the C-suite language.

Speaker4: [01:36:20] Their language is, I spend cash, how? Where am I going to get my return? And if you run your team like a startup, the C-suite is going to respect that. The C-suite is definitely going to move you forward. So even if you're an early stage of your career, if you run yourself and your projects like a startup, that's step one. How fast can I generate some cash with what I'm doing right now? Is there something I could be doing for my team right now that would generate more cash? I should bring that up. I should talk to somebody about that. Hey, I could be if I did this, it would bring up more cash. And it's not really, hey, this is cool technology. It's. I can make you some more cash right here. You know, I see a small bag of cash. Can we go get it? And it's that type of hustle mentality. And I think, you know, because of your generation and your generation has the hustle mindset, you don't have the work ethic. And I think there's a difference that we have to start calling out. You guys have hustle culture, you have hustle mindset, you don't have the work ethic. And the two are different things. Hustle is completely different. And, you know, and it really takes late Gen X, which is what I am.

Speaker4: [01:37:32] I'm that late seventies Gen X to kind of be able to bridge the two. And your generation is going to be super successful in companies that are run by people that are my age, because we're able to understand where you're coming from and we still have that kind of boomer work ethic that's built into us. So we have something to pass on to you about resilience and anti fragility. And a lot of these concepts that are going to turn [01:38:00] your hustle culture and your ability to hustle into your ability to spin up a business of your own. And we're going to be I mean, you're going to see CEOs my age start showing up, whether it's late forties, because that's the cycle that we're in. We've had a whole bunch of people retire early, just get out of the workforce during COVID. They said, You know what, this is so 100% not worth it. And you guys are actually killing CEOs that are in their sixties like that traditional late fifties, early sixties CEO. You guys are absolutely destroying them. They don't know what to do with you guys. So your entire generation is just pushing them out and it's kind of forcing leadership. And that's what I was talking about. And that's the other thing that if you're young in your career, there are going to be gaps to lead, especially in technology.

Speaker4: [01:38:53] People don't want to lead. They don't want to leave their technology skills behind. But if we don't have leaders with technical backgrounds, our companies aren't going to move forward. And so if those are that's everything that I would hit. It's that hustle. You don't become a worker or a drone because you lose what's valuable about yourself. Stay, hustle, hustle and go after that next job. Go after that next promotion because you're hustling for value and you expect value back. And that's the difference. The drone mentality, the worker mentality that was I'm just going to work and work and work and someone is going to magically give me, you know, it's going to come to me because it's owed to me and that's just not how it is anymore. And so you have that healthier hustle. Don't lose that at all. Whatever you do, no matter what people say, that's bad. No, keep the hustle loose. The the work ethic, the it'll come it'll come around to me someday. Become entrepreneurs, figure it out. Value, hack value quick go into leadership Find your way through the leadership find your way through and strategy and really the aim the end [01:40:00] game you're starting your own business spin off, start your own business get bought by your old one. It's you know, it's the recipe.

Harpreet: [01:40:10] Then. Thank you so much, everybody sharing a lot of great advice here on pretty much how to be future proof, how to be indispensable, where to take your career or what to do with yourself in your career. A lot of advice. There's so much. Let's go ahead and wrap it up. Be sure to tune in to the episode that was released today, a day later. Next week, I've got an interview with Dr. David Spiegelhalter, author of The Art of Statistics. He's also been on the BBC numerous times hosting shows on statistics like the things called The Forgotten and the show. But he's like three or four of them. It's amazing. He's a professor at the Cambridge University or University of Cambridge in England, one of those prestigious universities. He's there, had a great conversation with him, drinking a beer while talking to me, which I thought was pretty damn cool. Then after that interview that's released with Nick saying, We want an only mixing rock or pasty, so be sure to check that out. And then I've got just like four or five more brand new episodes being released, and then I'm going back into the archives and pulling out old ones because I've been doing this thing for I don't know how long. A lot. It's just been a lot of 245 published episodes in just over two years. There's a lot of content back there that I'm going to be releasing because I'm out of a lot of new shit to release because I have not recorded a podcast episode since early December.

Harpreet: [01:41:53] So that's why I was recording the actual interview, but I had that much of the backlog. Hopefully my basement gets fixed soon. I get all [01:42:00] my equipment back and then I can go back to recording podcast episodes, but I will be going back to the archives, just pulling out some good stuff that I know. There's probably only one or two of you that have listened to every single podcast episode. I know MTV is definitely one of them, probably the only one. I don't know who else is. If you have listened to all of my episodes, please send me a message. I want to know who you are. You deserve a virtual hi fi from me and I just would love to hear your feedback, but we have 245. That's a lot of work. That is a lot of work. That's it for this one. Take care of the weekend. Have a good rest of your day and listen to this on a weekday morning recorded. If you're going to be in Toronto on June 9th, 10th or 11th, I'll be at the conference with my very first live talk live demo representing package and shout out to the other conferences for sponsoring this episode. You could register for the conference as a link in the show notes, discount or preach 50% off and we'll take it. Guys, thank you so much for being here. I appreciate you guys. Take care of the rest of the day. Number one, I'll try to do something for.