FinOps in Action

FIA - Zak Ali
===

Zak Ali: [00:00:00] we still need humans in the loop. I think that's where our finops practice is still important. You can't. Maybe in a few years or the technology's changing every day, but I think there's still value for humans here. I went, you're wasting a hundred dollars. They go, I don't care about a hundred dollars. But they don't understand the compounding effect of a hundred dollars. A hundred dollars a day is a lot of money in a year.

Intro: Welcome to finops in Action. I'm your host, Alon Arvas. Each week I'll sit down with finops experts to explore the toughest challenges between finops and engineering. This show is brought to you by 0.5, empowering teams to optimize cloud costs with tip detection and remediation tools that drive action.

Alon Arvatz: Hello everyone and welcome to another episode of FinOps in Action. I am really looking forward to the conversation today because today we have a guest with over 20 years. Of experience in technology. He actually led his organization to execute a cloud first [00:01:00] strategy as a and as a chief architect of cloud, he developed a new operating model for cloud adoption, which I would love to hear a lot more about Today.

I'm happy to welcome Global Lead for Cloud finops and Cloud Modernization at Invesco us Zak Ali. Zak, how are you today?

Zak Ali: Hi. I am great. Thank you for having me.

Alon Arvatz: Yes. It's so great to see you and Zak. You have so much experience in finops and beyond. Please share with us what is the one thing you got wrong when you started your finops career?

Zak Ali: Yes, a very dangerous, uh, thing that happened to me, which was auto around automation, uh, the real dollars behind it, but not keeping my eye on the ball. let, letting automation make decisions and letting it run away and not putting any real governance around it, like humans in the loop. that's a tale for AI as well, right?

'cause it's the next stage of evolution. Is it? Give it. [00:02:00] decision making capability. But yes, that was probably my biggest, uh, when you sort of pose that question to me, I'm thinking, okay, yeah, that was the worst.

Alon Arvatz: Wow. Zak drop the mic. Like everyone talks

Zak Ali: Yeah,

Alon Arvatz: more ai, more automation, you know, don't engage with engineer, like stuff like that. And give us some more details. What did you try to automate and what did go wrong?

Zak Ali: Yeah, it was, uh, reservations, savings plan, usual. Right. Um, it wasn't our, it was a combination of automation and our, uh, tooling that we'd, uh, integrated. I think it's, uh. You know when, when you are, we give it the business rules and the our strategy and then that broke somehow. And in that it just kept on, you couldn't stop it.

There was no circuit break. It just kept on stopping, uh, which kept on buying, you know, all these reservations and contracts essentially we had to backpedal all of that and there's no getting out of it with the cloud providers, right? You have to go and start [00:03:00] striking deals and trying to mitigate that risk.

So that was. A couple of years ago when, uh, self automation came up in its maturity, uh, yes, we learned, we learned from that.

Alon Arvatz: Yes.

Zak Ali: still, we still automate, but we tend to do those things uh, by hand nowadays.

Alon Arvatz: Yes. It's very interesting and it's also fascinating to use the word governance. Typically, when people say governance, they talk about deploying guardrails to prevent people from making mistake. So you're saying no, you need to have people to prevent your automations from making mistakes, right.

Zak Ali: Right. I think it's human nature, right? So this is what I always think about. When give up to any type of technology that makes your life easier, you become lazier. We do it all the time. Uh, what phone numbers can you remember? How do you do math? [00:04:00] You have to pick up a calculator. Just basic things.

You, you slowly, uh, give that up and then those core skills, um, you end up trusting the technology that it's correct and it's not always correct. I think that's, uh, that's why we still need humans in the loop. I think that's where our finops practice is still important. You can't. Maybe in a few years or the technology's changing every day, but I think there's still value for humans here.

I don't, there's a lot of fear about taking our jobs away. don't actually think so. I don't think it's there yet.

it's a not a real promise. I think it helps and it augments humans and it makes you a more productive. But I think you still have to keep your hand on it.

Alon Arvatz: Yes, I totally agree. And you talked about it in the context of, uh, commitments. Uh, do you feel the same in the context of, uh, optimization from waste perspective like moving.

Zak Ali: uh, that's another one that was actually [00:05:00] earlier. So we had something, we would have business partners who were tech, you know, uh, when data scientists came up. So they wanted, they were like citizen developers right? Before it was technologists would. execute and put guardrails in place and all that sort of stuff. And then this new breed of professional came where they would code and they would actually deploy. So, but they had poor practices. You know, they had no idea, they had no thought of dev test prod. what happened was, uh, we would have this sandbox environment with automation in place with Lambdas running and things, and it would, we would allow people to do POCs.

Well, what actually happened was. There's a whole bunch of people running production in that environment and the automation came and just deleted everything.

Alon Arvatz: Oh my God.

Zak Ali: But that's not for me. That's, and that's really communication, you know, educating good [00:06:00] practices. I've done this most of my career. I'd normally come in, um, and we talk about where I came from. Um. I've interacted closely with the financial services traders and, quant analysts and all these people that, uh, very smart.

They pick up technology and start using it for business and deliver value very quickly, but then they start introducing risk, and that comes to where governance is, right? They start introducing risk because they're running, they're their compute under the desktop, for instance. And, uh, I don't think they understand what value technologists bring when they start looking at the world like that. But the cloud outages that occur, people think, oh yeah, this stuff's gonna stay up forever. It doesn't,

Alon Arvatz: Yeah.

Zak Ali: a bit more, it is resilient, but things happen. The Amazon outage watch, what was it a month or two ago, was major, right? And I think it really highlights, uh, highlights governance and risks and all these sorts of operating, [00:07:00] uh, hygiene that technologies have put in for years.

Alon Arvatz: Zak, what are you saying is music to my ears? I have to tell you, uh, because I feel always that things are a lot more complicated. Then they sound and everything is context driven. Like every little piece in your cloud environment, you gotta have the old context to make decisions. And transitioning that to automation is such a high risk that most of the time it's not worth it.

And we actually should strive to, yeah, automate things, but have engineer in the loop. Have a person in the loop. In the loop. So basically he can do some kind of governance. To what automation is designed to do.

Zak Ali: Right. Yeah. I think you

always need checks and balances regardless. Right.

Alon Arvatz: Yes. No, that's, that's great. That's important topic and thank you for bringing this up and also sharing, uh, your personal experience around that.

Um, and if, if we're talking about, you know, [00:08:00] implementing a finops, you know, across the organization, so you obviously gotta get the engineers engaged. So we talked about the people that have to be on top of that, have to do the governance. How do you drive this engagement?

Zak Ali: lemme just put things in context again. So, for Invesco, we have a Showback model. We don't charge back at all. I manage the entire budget for cloud, the, uh, infrastructure service that we provide, and any other services. And we have some platform services as well. So that's the sort of scope of my group. And we do the typical thing, right? So we've gotta shift accountability. We knew that that was by, when I wrote the finops strategy, it was the biggest, uh, and it came up in the state of finops many years ago. Engineers don't take action.

Alon Arvatz: Yeah.

Zak Ali: Well, I don't actually to totally believe that because they're doing stuff for the business.

So that's a totally different conversation we probably could have. But you start with the boring stuff and um, [00:09:00] then. Just showing them what they're spending and then interacting with them and helping them. And now I've got an architecture background and engineering background, so it's easier for me can talk about best practices, uh, cost design by design for cost, make sure that it's a continuous improvement exercise. It is overwhelming for a lot of engineers that go to cloud because this is a new function that they have to take on. They don't understand. They think it's me, and I'm going, no, it's not me. It's you because I'm not doing anything. And that's what used to happen. People used to go, why are you spending this?

I'm going, I, I'm not spending anything. I'm just, I'm just showing you. And I think go, showing that Showback model and showing them, know, the journey gets you so far, but. The challenge we had is that in a showback model, they're not really accountable or responsible for the spend, right? There's, nothing's gonna burn them.

They're not gonna get an invoice. We get the invoice. So we're constantly chasing them. So what [00:10:00] what we did was, um, every year we define objectives and key results, just basically goals, um, under goals underneath these big numbers. So we, we have a big number that I set and we say, okay. Through optimization, through the finops practice, through waste and savings.

We're gonna save, we're gonna go after this number. And it's avoidance, right? Because it's not really savings. You're effectively avoiding the cost up until the end of the year when you sort of rationalize the number and reconcile it. So, um, a couple of years ago, probably, yeah, probably a couple of years ago, um. We had this really big number that we couldn't get to, and it's because we were trying to push others to go and and reduce waste and just be good citizens, create standards so we can buy reservations and things like this. Uh, so I sat, uh, on my couch one day looking at one of our stories. I'm thinking, okay, how do I make this [00:11:00] fun? I went to my friend chat, GPT, and uh, I came up with gamifying. Um, this. activity. And what I did was, I called it Frugal Guru. I didn't call it Frugal Guru, uh, tragedy told me it was Frugal Guru. So we've got this efforts called Frugal Guru, in the Frugal Guru, I make everything fun. I even presented it to our technology group, which is everyone in technology, and I made an this entire presentation out of memes. And gave them big metrics like a couple of years ago. I think there was, I'm just, I'm probably making these numbers up, but just to give you a size of what I was conveying, I'll say, I'll tell you why. I would say, Hey, know, cloud spend for 2024 was $160 billion, and of that $160 billion, $46 billion was wasted.

[00:12:00] And everyone go, wow, that's a big number. Because I normally go to a an engineer, I went, you're wasting a hundred dollars. They go, I don't care about a hundred dollars. But they don't understand the compounding effect of a hundred dollars. A hundred dollars a day is a lot of money in a year.

Right?

Alon Arvatz: Yeah.

Zak Ali: they

don't see. So then I drew them in and then I gave them this presentation. I gave them basic things to do. So park your instances. Autopark. So I had to Tesla Autopark itself, and then I had, uh, the sandbox environment because we stopped a lot of automation. I said, Hey, just get rid of your stuff.

Just be a good citizen. Stop, you know, if you've done your experiment, delete your instances.

Zak Ali. : So like all presentations, you keep them simple, right? You give them one takeaway, you give them three takeaways. So I gave them three takeaways. The first takeaway was try and auto park your instances apart from production. Normally you can auto park everything, right? No one works 24 7. So I [00:13:00] gave them not only the action to autopark, I also tell them the effect.

So if you didn't auto park. Let's call it in Amazon. We are mostly in Amazon. We've got. Our primary landing zones of Amazon got a big uh, environment there. So I would say, Hey, park your EC2 and if you don't park, this is how much it would cost you in a year. And if you do park, this is how much it would cost you in a year.

That's significant savings, right? The weekends after work, et cetera. And then the second thing I did was clean up your environment. So when they've got these sandbox environments, which you can't regulate, well, you can't really tag them well, you can't do anything 'cause there's no real governance. It's all about experimentation.

But there's a cost to that. And it's significant because people were just leaving things on and then we stopped that policy 'cause of my mistake. Uh, and, um, we told 'em to do that. And then the third thing we asked them to do was optimize. And I had this, I would find these memes on the internet and there's, [00:14:00] there's a guy, and you know, those, uh, Tokyo Hotel capsules.

Alon Arvatz : Yes.

Zak Ali. : You know, you know what I mean? Right. You could the one, yeah. So this guy is drumming at the, in a capsule like this, I went optimize your instances, and then I, I would give him examples of a database if you optimized it and if you, you know, optimize the server instance, et cetera. So I did that. So those are the takeaways.

And then I finally finished. I said, well, I'm not gonna take credit for your hard work. And what I do, I, I just made it fun because this is what I do. Everything's a comedy show. So I gave them. When, when I recognize them after, so they, what they would do is actually submit, uh, what they saved and we would then, my team would then validate what they saved and then we would give them a million dollars.

Alon Arvatz : Nice.

Zak Ali. : And I think there's a, there's a lot of conversation about how do you compensate people. I think people just wanna be recognized and I think the most important thing is. And I say it a lot, a lot of my [00:15:00] colleagues, and as you get, as you get to know me, I'll say everyone understands what a dollar is. You start talking technical language, no one.

But if you say, Hey boss, I saved $4,000 because I cleaned up some storage. They understand that, right? They might not understand the technical value you deliver, but everyone understands what a dollar is. So that's how I drive it. I don't really talk about technology. I talk about the dollars, and then make it very simple.

And that's what I've done with the executives as well, and we probably can talk about that and how I get them to take action, uh, which is another strategy I take. But that's what we do, and I think it solved two problems for us. One is that it got them to take action in a fun way. You know, we didn't put a lot of pressure on 'em.

And then they understand by, you know, even if you save a dollar, it's a, it's death by a thousand cuts, right? All these dollars then add up to big dollars. And then the second thing was that when we now, um, [00:16:00] the accountability shifted. We didn't get those insights. We didn't know who was saving what. That was another problem, right?

So they were now coming back to us, they were closing the feedback loop. So it works both ways for us. We still run that to this day. I think it's a couple of years old, this effort, but, uh, it

works. And oh, that's what we're.

Alon Arvatz : that's a safer automation for sure. Uh, wow. Zak, you're not a, a cloud or finops leader. You are a marketer, basically.

Zak Ali. : Yeah, I think, uh, these are all skills that you learn now. A long time ago, I forgot who it was now, um, but originally when the foundation started, there was this slide and it showed a unicorn. And it, I dunno if you remember that, but it was a unicorn. And that finops unicorn interacts with all these people.

And I think that's, you have to sort of develop these skills, right? Um, for instance, there was one where. In fin practice, you [00:17:00] need to hire a designer. I'm thinking, why would you need to hire a designer? I don't even understand. And then it comes to light. As you progress your maturity, you, you escape out of the insights that's done, right?

All the reporting's there. You've got all the technology in the world. You have to, you have to develop these other skills like communication, marketing, because this is a very boring topic as well. Right, and people shy away from money in the first place, so you have to make it a bit more interesting and it has to have longevity.

So really liked what you said in the beginning that people say that engineers don't take action. It's not true. They take action, but the take action, what they see as a priority, if they don't understand the impact, for example, the compounding impact. Of their actions, or if they don't understand, you know, what are the implication, what they do?

Alon Arvatz : Of course they won't take action with things related to finops. They'll just take action in other areas. So it's really about how you [00:18:00] make it a priority for them.

Zak Ali. : what's important there and what I've been doing is explaining the macro effect. I, I'm working at. 50,000 foot. Right? My budgets are really big. Um, the money that I see is a lot of money. The commitments that we make is a lot of money and the engineers don't say, see that They've got like a sort of small stack, uh, environment that's few hundred bucks.

They don't understand the impact and it is that taking them away from that and showing them the big numbers and then they, then they feel like they have responsibility. And I think, I think my, um, our global head of infrastructure said it once. He was really surprised that they took action. He surprised him, but I said, everyone wants to do the right thing.

Everyone's a good corporate citizen, right? I don't, I think that's just human nature and people wanna help each other and you have to give back as well by recognizing them as well. You can't just take all the credit. So it's

important.

Alon Arvatz : Yes. And I feel it's also a very important [00:19:00] takeaway that in the end of the day, what they're looking for is recognition.

And you like giving them like these million dollar bills that you created that's giving them this recognition. I am curious though, if you also consider giving them like financial incentives, like, uh, gift card or like, or budget for cloud, whatever it may be.

Zak Ali. : yeah, we, we thought about those things, but. I call businesses ebbs and flows. So if you set that type of precedence in times of where we aren't so, um, flush with money, uh, it is hard to do, right? So that I think you set an expectation. There's also the gamification of this. So people start collecting lots and lots of dollars and then you realize that they're not really following the principles because there's real dollars behind it.

So I think. It may work in certain cultures and certain, uh, [00:20:00] uh, not certain cultures, but certain business cultures, uh, depending on what your company and how it's driven and how you are compensated. Um, it's certainly something that I didn't want to go down. I think that's too difficult to do and manage, and then you, you're messing with a lot of real dollars and I think this is just as effective enough anyway, so.

Alon Arvatz : Yeah.

Yeah. I, I think, yeah, I think if you just give them the opportunity so you, you, you save the most money, get them in front of a C level or people where, where I have that ability. I don't always recognize that someone, you know, an engineer, a junior engineer, would love that opportunity more than being compensated, right?

Zak Ali. : Because they're looking at progressing their career. I'm on the tail end of my career.

Alon Arvatz : Yeah, for sure. And I, I think it's also key to make sure they understand that. Finops and finops skills, advances their careers, advances their skills so that they become better engineers, where they're now getting more [00:21:00] proficient on costs and efficiency.

Zak Ali. : right.

Alon Arvatz : Okay, great. Uh, and I have another question and wow, I have so many questions, what you said, which is great.

You mentioned about the distinction between cost avoidance and cost savings. I wonder if you feel that this distinction is helpful when, when it comes to one, communicating with executives and two, communicating with engineers.

Zak Ali. : No, I think it's more of a financial thing, so you could use it a different way. You could say, okay, I've got a budget, the budget's, the budget. If I say that I'm saving cost, is someone gonna come to me and say, I want that money. Finance typically do that, right? They reconcile every quarter, they take their savings, and it has a net effect on the profit and loss on the p and l. But if then I use that with engineers, I don't think it makes much difference to them. You know, the, the effect is [00:22:00] different. The cost avoidance is good. Because then the real dollars stay static. So you can use those. So it is similar, I think, I don't sort of, I know where you're going, not that I've done this, but you could reinvest those avoidances, right?

You could go, okay, hey, I want to experiment. I'll invest in your innovation and here's some dollars that I can put to it because I'm not blowing my budget. I'm underrun my budget. So I think that's a maybe a good strategy or a good tactic to use. From with cost avoidance, because if you, if it was cost savings, you may not even have the dollars to give, give back.

Alon Arvatz : absolutely. Okay. And. Obviously it sounds like you really, you know, worked with engineers to build this motivation, awareness, gamification. I heard a lot of times on the podcast that actually like doing things top down first, eh, is the right way to do that. I wonder if you share the [00:23:00] same perspective or no.

Like it worked for you straight with engineers and it's not necessary to go

top down.

Zak Ali. : in our business, again, it's always contextual, um, because we're showback, it has to be both. So the way we did top down was. I, well for, I'll tell you the journey 'cause we've had many finops tools and create all these dashboards and management dashboards and kept on going. I don't even wanna think how much money I spent building those dashboards, but they were never affected.

No one actually logs in. And then one day, CIO at the time said, you know what? And because most of our budgets are show back, um. He wanted to show the business, like what effects they're having on technology spend. So he says, I want to invoice them, but like a fake invoice. Oh yeah. That's interesting. [00:24:00] So what I did was, um, I went, well, cloud's a utility, I'll use my utility bill.

So I took my. Um, utility bill, if anyone's been in any of the finops events, they've probably heard some of my, uh, Bnai, she's actually presented this and I've actually presented this idea. So we call it a cloud utility bill, and it's a one pager and, uh, it has spend size like Euro, electricity bill in, in England, in here, in, in Texas.

Uh, I was with a company called, uh. Uh, I'm not promoting them anyway because I've left. But typically, I'm not sure what your utility bills look like, but you get a one pager, right? You'll say, Hey, or your cell phone bill, whatever it is, it'll say, this is what you spent, um, this month. Uh, this is where you've spent it.

And then especially with power or any sort of utility, they'll give you ideas of how to save money as well. They'll say, Hey, turn down your thermostat one degree. And then it also [00:25:00] shows you, here, it shows me what my neighbors are spending generally. People in your neighborhood spend this. And especially in Texas, it gets hot.

So when my kids were at home, they'd just, this room here was my son's room. He would just, it'd be like a fridge, right? And I'd run after him saying, Hey, why, why are you spending this much money? And, um, I think I used the same concept with this utility bill. So I got, this is where the designer came in. We took the electricity bill, gave it to the designer.

The designer marked up a cloud utility bill, and then we built. We built an infrastructure app. It's not, it's fed by a lot of, um, it's fed by the cloud provider data, the curve files, um, the, the Azure billing data, um, and the other cloud providers. It's also fed by our finops tool, uh, with all the context from our CMDB.

So now we are shifting it from cloud data into business data. So what is our, we're an asset. We're in a, [00:26:00] uh, financial asset management. So what's our investments team spending? So now this bill, we produce multiples this bill, which has how much you're spending, how much your neighbors are spending, so other businesses, what you spent last month, what were your top spends.

And then we have this thing called Smart insights, which was, it's somewhat prompt driven. Now we're trying to put AI into this, uh, 'cause it's very difficult to do. You know, there's only. Three of us in the, uh, four of us in the finops team. So there's, you can't scale this, but we are trying to scale it. Um, it tells them very basic things to do.

Like an executive could read it, like your service spend is this, or, uh, this environment, which then they correlate to the business is doing this. And then we give 'em very simple language to take action what they do. And then behind that is another dashboard which you can drill down into and then find, it goes to the finops tools, which the engineers can look at.

So [00:27:00] I'll just tell you the net effect. So, um, our, our. Head of investments technology gets this Bill gets the bill in his email, gets it, gets it in his email as a PDF, and in there it's saying, this group is spending, this business group is spending this, or whatever it is, or we are giving them ideas or to take action. He then takes it to his directs and says, do something about this, and then it trickles down. That was super effective. It got. It's such a, um, gem of an idea and, um, an asset that when we, we stopped doing it after a while, we slowed down because it was too, it was getting too popular 'cause people were sending it to each other and then it got, oh, my business wants that bill and my bus.

I'm thinking, okay, I can't do this anymore. So we slowed it down a bit. We're speeding up again. But, uh, that, that head of was taking it to their business head of. [00:28:00] In a meeting and saying, you know, this is what, um, this is your cloud consumption and this is the effect on this strategic project, for instance, or on your business.

So it's very effective. And then we collaborate with them. So I've got another designer and data engineers, and now we recently, we've been enhancing it first 2026, where I've got lofty ideas of where this thing's gonna go. But that's been super effective. I know it is long and drawn out, but. I would say that I haven't found any tool that can, you know, replace that, right?

Because there's so much institutional knowledge there. That's something that my team, you know, delivers uh, a upmost value.

Alon Arvatz : That is so creative. Oh my God. Well done. And uh, Zak, well we already got so many insights from you, but I know that you also have a lot of experience and a lot of interesting points to mention around communicating with executives. [00:29:00] So I wonder if you can give us like, you know, 1, 2, 3 tips on how to do it right.

Zak Ali. : Uh, first of all, I've said it already, which is simplify the language. No techie speak, talk in dollars. And then the other thing is to understand their business. And most of my team, they know this, probably hate me for it. Is anyone who starts in my team, I a I ask them to go and understand the business because as technologists, and I think more and more over the years, um, and I came from a different field, we probably can talk about that, but, um, I recognize earlier in my career, it's important to understand how a business operates.

Because going back to finops, let's say your customer's spending a million dollars, is that good or bad? You can't answer that. If I ask the cf, if I ask the CFO they'd go, it's bad, switch it off. If I ask the business, they go, no, that's nothing. We are, we're bringing [00:30:00] in hundreds of billions of dollars in revenue.

$4 million. Why would I go after them? I, I wouldn't. Right? So, uh, I think that's what's, they're the three things, right? Contech understand, uh, simplify the language. I've forgot the third, second thing you're probably gonna remind me now. Uh, but the third thing is really understand the business. I think that's super important.

You have to do that.

Alon Arvatz : Okay. That's awesome, Zak. Wow, I, I got so many amazing insights, limitation of automation, importance of governance, gamification, engage, like all of it. I loved it, but I'm not letting you go before I learn more about the personal

level, Zak, it's not over until it's over. So first of all, where do you live?

Zak Ali. : I live in, uh, as you can tell from my accent, this is my always joke. Uh, I live in Houston, Texas. Uh, the only thing I've said is y'all, but you know, uh, so, uh, yes, I've been here since 99.

Um, I, [00:31:00] I worked in London, in the city. I worked for Reuters at the, uh, when there were a bigger company. Um, and I came from, um, market data.

So I used to. Work on trading floors. And that's how I actually got here. So I used to be a market data engineer. That was my, how they hired me at, uh, at Invesco many, many years ago.

Alon Arvatz : Okay, so how did you get to finops eventually?

Zak Ali. : Yeah, that was cool. Uh, that was totally passion driven. I think I had another podcast when we were talking and the person who I was talking to, they said that I was doing finops before finops was an organization and, um. The reason we got, we got into finops. Well, the reason I got to finops was that, um, again, it's that power of I know what the, why this is important and why we're doing it.

I knew the business reason we were going to the cloud and it wasn't because, um, the sell of saving money, there was no such thing. [00:32:00] What we did was we did, um, I did an analysis on. How much, uh, there was this small company that they're from, MIT, the group of guys, really smart guys. What they did was they developed an, a workload that they could deploy.

This was years ago, by the way. Uh, like in 2016 or something. I don't know. It was a long time ago. They would deploy a workload in our data center and they would deploy workload in the cloud. And we had two clouds at the time, and I would measure those. We validated that infrastructure as a service was not worth it because we just refreshed our data centers, so we were more optimized in the cloud. So we said, okay, so what can we use it for? Well, you know, we had all a bunch of data scientists started and they wanted Hadoop clusters and all these sorts of big things. And we're thinking, okay, well the cloud providers can. Faster, better and cheaper than us. Like I wouldn't want to go and build clusters in a data center that's too expensive.[00:33:00]

So that's how we used it. We used it for new capabilities. So from that, finops became naturally, I started measuring everything by dollars. As soon as we, you know, when I got my first $40 bill on my credit card, I was measuring it from then. However, finops didn't become a thing until the number got really big. Like I was still doing things 'cause I was still the architect, I was still helping people do things by, um, design for cost, but in a very small way. Um, but at the top they don't really care. We are, we were rounding era compared, and I think when it really became, ops became a real thing and I got. More staff was when it was the third largest cloud, was the third largest spend in our technology portfolio. People take notice

Alon Arvatz : Yeah,

Zak Ali. : what's, what's this?

Alon Arvatz : [00:34:00] absolutely. Absolutely. Okay. Wow. So you went to the US 99, uh, since then you're in, in the US and now Houston. That's great. And Zak, I always ask and I always find it interesting. Is there a book or a movie you would like to recommend to us?

Zak Ali. : Yes, my, I'm always, uh, I'm always reading, uh, either listening or reading. I think my current, uh, and it's obviously. On Vogue right now, ai. Um, but there's a, a book called Empire of ai and it's by Karen Howell. And she is a journalist. She's actually an MIT engineer, I believe someone can fact check me. Um, and she became a journalist and then she got embedded into open AI way back in the, when it was, um, still a nonprofit. And, uh, she talks about that journey. I haven't [00:35:00] finished it by the way, but I find it hu uh, super interesting. There's another, uh, book called Super Intelligence, which is similar. It's really about the ethics and, uh, how power and money has driven AI and some of the ethical things that have happened. You know, we, we are hearing it a lot now where, uh.

These companies are trying to build data centers and they're depleting resources and all of these things. So it's very an interesting point of view beyond, you know, the coolness of ai. It's just the net impact of this technology and the powers behind it. So it's interesting. We, I would, uh, highly recommend it.

Alon Arvatz : Very interesting. Very relevant as well. So something to add to my reading list. Cool. And Zak, what do you like to do for fun?

Zak Ali. : some people don't call this one, but I run, I run half marathons. I used to run full marathons all the time, but I'm running half marathons. Um, so I tend to do that a lot, and that was because I'm a super nerd [00:36:00] and became very unhealthy over my lifetime. So it's one of the things that I, I call it my meditation.

I just zone out and do long distance running. So that's one of the things I do. Again, I've got my nerdy side, so I've got lots and lots of hobbies and too many hobbies and many projects I've never finished. But, uh, I love, I love

running.

Alon Arvatz : Yeah, and that's the first thing you said today that I cannot relate to whatsoever. The first thing, so I hate running, but I love swimming,

Okay. That's great. And Zak, wow, so many interesting insights and experiences. I'm sure people would wanna reach out, ask questions, get some consultancy.

What is the best way to reach out to you?

Zak Ali. : Yeah, I'm on LinkedIn. You found me on LinkedIn. Um, so yeah, go ahead and post my profile. Happy to chat. Day and night about finops, and obviously

I've got a lot of experience around cloud modernization as well, so happy to

share that.

Alon Arvatz : Absolutely, and, and thank [00:37:00] you for sharing all of that. Like it was all really insightful. And I have to say also really practical, like how you take your engineer step by step towards getting awareness to their ac, the accumulative impact of the actions, showing recognition, driving, uh, adoption, driving motivation.

I loved all of it and I'm sure that our listeners as well. So Zak, thank you so much for joining us today.

Zak Ali. : Thank you and happy holidays.

Alon Arvatz : Yes, happy holidays. And actually there is a good chance it'll be published after the holiday, so I hope you all had happy holidays. Um, and with that I also wanna thank our listeners for joining us once again.

Please tell your list, your friends about phs in action, how much you like it. I wanna have more people aware of the podcast. Zak, thank you again. This has been another fun. Honestly, I had a lot of fun today, so it was another fun episode of FinOps in Action. Hope to see you all [00:38:00] next time.

Outro: That wraps up another episode of finops in Action. Thank you for joining. For show notes and more, please visit finops in action.com. This show is brought to you by 0.5, empowering teams to optimize cloud costs with tip detection and remediation tools that drive action.