DevOpsify with Sasha Rosenbaum of Ergonautic

About the guest:

Sasha Rosenbaum is Principal at a new venture, Ergonautic. With a degree in Computer Science, an MBA, and two decades of experience across development, operations, product management, and technical sales, Sasha brings a unique perspective to optimizing the organizational flow of work, bridging gaps with empathy and insight.

Find our guest on:

• Twitter
• LinkedIn
• Mastodon

Find us on:

• On-Call Me Maybe Podcast Twitter
• On-Call Me Maybe Podcast LinkedIn Page
• On-Call Me Maybe Podcast Mastodon
• On-Call Me Maybe Podcast Instagram
• On-Call Me Maybe TikTok
• On Call Me Maybe Podcast YouTube Channel
• Adriana’s Twitter
• Adriana’s Mastodon
• Adriana’s LinkedIn
• Adriana’s Instagram
• Ana’s Twitter
• Ana's Mastodon
• Ana’s LinkedIn
• Ana's Instagram

Show Links:

• Ergonautic
• DevOps
• DevOps Days Chicago
• FileZilla
• [Video] “Single Person of Failure” by Sasha Rosenbaum
• Google SRE Book
• Andrew Clay Shafer

Transcript:

ADRIANA: Hey, y'all. Welcome to On-Call Me Maybe, the podcast about DevOps, SRE, observability principles, on-call, and everything in between. I am your host, Adriana Villela, with my awesome co-host...

ANA: Ana Margarita Medina.

ADRIANA: And today we are talking to Sasha Rosenbaum, who is a Founder at Ergonautic. Welcome, Sasha.

SASHA: Thank you. I'm so happy to be here. And it's a pleasure to be chatting with you all on this podcast.

ADRIANA: Yeah, we're super excited to have you. So first things first, what are you drinking today?

SASHA: I'm drinking peppermint tea. And this is a funny story; I recently became allergic to black tea.

ADRIANA: [gasps] Oh no.

SASHA: Which I previously drank all my life. I didn't even know you could be allergic to black tea. So now it's either coffee or I try to find some type of herbal thing. [laughs]

ANA: [laughs]

ADRIANA: Oh my goodness. At least it sounds like you've found a nice alternative. 

SASHA: Yeah, I have so many variations of herbal teas. It's not even funny. 

ADRIANA: [laughs]

SASHA: Because I keep trying them all, and I can never quite find one that I'm committed to, so it's just a whole big cabinet of tea.

ANA: [laughs] I am sorry for you having to not have access to black tea anymore. I think that's always been my favorite, like, oolong tea/black tea, and especially when you get to combo it with a peach or a mint on top, like summertime. But I'm glad that you can still drink coffee because I know for some folks, it's like they have to cut off tea and coffee.

SASHA: Yeah, it's funny because the hardest part is people don't usually have hot drinks that aren't tea and coffee. So I'm like, if it's like 8:00 p.m., I'm not going to drink coffee, so I'm stuck with water.

ANA: [laughs] Fair enough.

ADRIANA: Yeah, that's sad. That's sad.

ANA: I'm very close to Sasha's drink. I'm having something called a mint mojito. There's a coffee shop in the Bay Area called Philz Coffee. And this is one of their most common drinks where it's like you get a latte, but they also muddle mint into it, and they make it really sweet with a lot of foam. So that was kind of what I was feeling this morning.

ADRIANA: Nice. Nice. I did not go with anything creative. I just have my water here.

ANA: And it's too early for boba tea for you.

ADRIANA: I know, right? Yeah, I think the places don't open until at least 11:00 if I'm lucky. Otherwise, I have to wait till noon or make my own. [laughs]

SASHA: Oh no. Oh, can you make your own? I've never tried making my own boba tea.

ADRIANA: It's not bad. Like, you can get the pearls on Amazon, and you just boil them for like one or two minutes. And if you want, you can soak them in honey or brown sugar, and it's actually pretty good.

SASHA: Nice. You know what I like? The new ones are like jelly, so lychee jelly or something like that. And it's just like, I like it more than boba because boba is so giant. [laughs]

ADRIANA: Yeah, that's true, and it's filling too. They pile so much on. I'm like, oh my God, I'm like eating my drink. [laughs]

SASHA: It's a whole snack.

ADRIANA: Yeah, exactly.

ANA: I'm insane, and I like it when it's boba and then lychee like jelly. It's just like the double whammy of it.

ADRIANA: It's a good combo.

SASHA: Any drink that has lychee in it, I'm all for it. [laughs]

ADRIANA: Me too. Oh, I'm such a huge fan. Whenever I find the fresh lychees at the supermarket, I'm like, oooh, give me, give me, give me because they're so, like, it's seasonal here. I don't know how accessible it is where y'all live, but here, I think I can find it in February and maybe one week in the summer.

ANA: We definitely don't have too much access to it here. I think I had more access to it in Miami and in Central America. But I want to say in Downtown San Francisco, most shops always carry lychee, which is actually pretty neat because I'm like, I prefer this over anything else.

ADRIANA: Oh yes.

ANA: Or when you go to froyo shops, they have boba, but it's the popping kind, and it's got little like bursts.

ADRIANA: Yeah, that's right.

ANA: Then, at Target, I just saw that they're selling pre-made boba, and it comes with those like popping boba. So I bought it for my partner's kids, and we've yet to have it.

ADRIANA: Oooh.

ANA: But it comes in a little four-pack pre-made boba, and I was like, that's fun.

ADRIANA: That's awesome, little family activity.

ANA: Yeah. [laughs] For taco night, my plan was we'll do tacos, and then we'll have these as desserts. [laughs]

ADRIANA: That's awesome. [laughs]

ANA: Well, Sasha, we'll actually get started on some of the questions that we have for you today as much as we love talking to you about your love for herbal teas. [laughter] The first question that I'm sure some of our listeners are wondering, like, how is it that you got your start in tech? We know that you've gotten a chance to work at various companies. But it's always nice to hear how folks were like, ooh, this seems interesting. Let me take a look.

SASHA: So I'm actually, like, I'm probably at this point an unusual woman in tech in the sense that I've been here almost 20 years. And I actually have a computer science degree, so it's like a more conventional path rather than switching careers. But I did actually switch. So I was getting the same message in high school as everybody else that computer science isn't for girls and that kind of stuff. And so what I initially enrolled in university was biology, and I wanted to do genetics. And the first year in biology, basically, I realized the entire degree is like growing yeast and such things in a lab. [laughter]

So you do genetic experiments, which sounds really, really cool. But to modify these genes, you grow fruit flies, and yeast, and then mice, and stuff like that. And I'm like, you have to proceed to like Ph.D. to like...and the whole time, this is what you do. You show up every day to your lab, and you're like, feed mice something. [laughs] And I'm like, this doesn't sound like a career I want to have.  

ANA: [laughs]

SASHA: So I kind of went...and my university was extremely technical, so I went through the entire list. And it was like, chemical engineering, physical engineering, industrial engineering, this, this, and this. And then there was computer science. And I did one Pascal class in school that was really basic Pascal, and I was like, I kind of like that. So this was like, click a button, switch to computer science. But then the moment I was there, it was so fun because it was like basically playing games.

What I love about computer science is you start with nothing, like, this blank slate, you know, empty editor. And then you create something that actually produces results, and that is just magical. It's super creative. You have a lot of freedom in how you structure things, and what language you use, what architecture you use. So it was always like, super, super fun to do. That's how I get started. 

And then, I was a developer for many years. And then I gradually got into...I did some ops in the beginning, and then I basically got into the DevOps movement. So that kind of shifted my career towards more opsy, SRE kind of conversation. I think I've held every role that you can imagine. So I was a product manager for a minute. And I've been in sales. I've been in DevRel, which to some people, it's marketing and to some people, it's not. So basically, I feel like I've been in every department of the organization but always something tech-related.

ADRIANA: You talked about getting involved in the DevOps movement. I wanted to dig a little bit deeper into that. And specifically, how have you seen the DevOps movement evolve since you jumped on it? And follow up to that is are you happy with how it's progressed?

SASHA: I got involved...I think it was 2013. So technically, the first conference was 2009, so it was probably about four or five years later, but it was still not a thing. So when people talked about DevOps, like, if you look...there's in Google search results how does search for a certain word change over time. And so the search for DevOps was, like, from 2009 to 2014, it was near flat. And then it kind of starts spiking because people got more and more interested in the term. The devopsdays.org kind of spread the word and started popping conferences all over the world. 

So I was a speaker at DevOpsDays Chicago, I think, in 2013, and then I became an organizer in 2014. And in 2013, it was the fifth city that did DevOpsDays, and now there's like, I don't know, over 80, something crazy. So it's all over the world. It's everywhere. So I think there have been a lot of changes in how it started, and how it progressed, and where it is now. 

What we started with was basically a lot of talk about collaboration and a lot of talk about automation. So it was two things, like, one is aligning incentives, so you stop kind of chasing two different agendas. Basically, you know, and I'm sure your listeners know, that developers want to deliver faster, ops want to keep the lights on, so minimize change. And then there's this inherent conflict between the team.

So the first conversation in DevOpsDays, how do we minimize this conflict? How do we make everybody speak the same language, understand that they're incentivized to do the same thing, which is delivering software to their customers successfully? Then it evolved to people started creating roles for DevOps, which at the beginning was a big no-no for us. I had one of my co-organizers refuse to take a job at a company as the director of DevOps until they renamed the role because he was like, DevOps is not a team.

ANA: [laughs]

ADRIANA: Yes.

SASHA: And I'm not going to work here.

ADRIANA: [laughs]

SASHA: We have definitely lost that fight. DevOps is now a team.

ADRIANA: Yeah, that makes me cry.

[laughter]

SASHA: So I will say on the bright side, the DevOps teams we did kind of create a job. So, in addition to...because DevOps teams don't exactly do what ops team did and they don't exactly do what dev teams did, so it's kind of like this new set of skills that were defined by this movement. The DevOps term also got co-opted by everyone, and their, like, every company selling a product in this space wanted to say they sell DevOps.

ANA: [laughs]

SASHA: And so every day I read some article saying that DevOps is dead [laughter] and usually to start an argument. Usually, someone defines DevOps as whatever they want, so they're like, DevOps is all about culture, or DevOps is only CI/CD tools, or DevOps is this. And then this is why it doesn't work. It's like, okay if you define the word to be whatever you want it to be, you can argue against the concept and practice. And then I think the first couple years that SRE became popular, I was kind of like, oh, you're just renaming this thing to another term. But I've come around because SRE actually defines metrics that allow you to reason. 

So we talked about incentives, and the incentives are great while you can talk, while you can collaborate, and everybody can be in the same room, and they can communicate with each other. But over time, in the larger organization, that's kind of hard to maintain. If you can attach metrics to it that allow you to reason about this conversation, it makes it easier. So SRE did bring a lot to the table as, in practice, you can implement with SLIs, SLOs, and error budgets and all these things that kind of allow you to discuss what the trade-offs are, which is really important. 

I think the unfortunate part of it is a lot of organizations just rename the team to whatever the buzzword is. So ops has been renamed to DevOps; it's been renamed to SRE. Now it's been renamed to platform engineering, whatever comes next. And they don't change any of the practices. So how do you expect to get better results? Like, you say you're SRE...one example is so many organizations are like, we are an SRE team. I'm like, okay, what's an error budget?

ANA: [laughs]

SASHA: And either they don't have an error budget at all. They don't even know what it is. Or they have a pretty slide saying error budgets are this. They don't have it defined. Or the next one is they have it defined, but no one looks at it. And if you blow the error budget, no one stops development, and they just proceed as normal. I'm like, so you're calling yourself SRE based on what exactly? 

ADRIANA: Right. [laughter]

SASHA: It's like, I'm monitoring alerts [laughter] like we did 20 years ago, oh no.

ADRIANA: It's like fancyops at that rate, right? 

SASHA: Yeah, yeah. There's a lot of fancyops around it. And so I had this moment, I don't know, three years ago, whatever, somewhere during or before the pandemic. 

ANA: [laughs]

SASHA: I had this moment where I came to my co-organizers of DevOpsDays Chicago, and I'm like, "I feel like we didn't do anything. We're having the same conversations for ten years. Nothing changes. Everything is sucking in the same way." And they're like, "No, no, no, you're wrong." My job used to suck. Like, my job used to be I had a once in six months deploy. It took me all weekend. I was on-call for it at 2:00 a.m., and no one cared about my well-being. And developers were like, "We don't care about operations and whatever." 

And now I get to implement tools and practices, automate a bunch of this, get releases every day, which makes it less painful, all of these things. So we improve the job of operations a lot. That being said, there are so many, like, if you look (I had it somewhere in my slides.), if you look at statistics from StackOverflow, they did a survey on deployment practices. And the last time they did it was 2017, and they stopped doing it, unfortunately. But I think in 2017; it was something like 20% of all companies deploy through uploading zip files. [laughter] I'm like, where are we? You don't have a single pipeline? You don't...what? Okay. [laughter] 

ANA: Like we bring back FTPs and just dragging and dropping like no sense of version- control apart from just dropping your code in some FileZilla. [laughs]

SASHA: I've literally walked into organizations where they actually didn't have source control. So they're writing software, and they actually don't have source control, and this was like two years ago. To some extent, it makes me mad when people are like, "Oh, DevOps is such a solved problem." or like, "Everybody's doing CI/CD and deploying everything." [chuckles] I'm like, no, this is not what's out there. There's such a different landscape. I love this quote, like, "The future is here, but it's not evenly distributed." 

ANA: Yes. 

ADRIANA: That's a good quote. 

SASHA: There's such a different landscape with people doing the best best practices of everything and then the worst practices of everything. And then I feel like the problem with emerging tech is people continue to try to sell the new, shiny thing. So an example...and I don't love picking on particular things but companies that sell chaos engineering products...this company can't even deploy code, and you're trying to sell them chaos engineering. [chuckles] Let's go back to the basics. Like, what are we chaos engineering here? [laughs]

ADRIANA: I do feel, though, in large companies, I mean, large and small, really, like there are VPs of tech or CTOs who fall prey to the shiny new object as well and then end up leading the company [laughs] in this path of like, let's implement this tool, and, as you said, but their house isn't fully in order. So how can you implement this tool when the foundation still sucks?

SASHA: Part of this is how people get budgets to do things. So part of our job at Ergonautic we do management consulting in IT. So we kind of come in and try to help organizations improve their IT practices. And one way to phrase it is like, we'll come in and rescue your DevOps transformation. 

ANA: [laughs]

ADRIANA: Love it.

SASHA: Because you failed, or you think you failed. And we'll come in, and we'll, like, you definitely improved some things, but you're probably hitting some bottlenecks. And we'll come in and just take a view of the landscape and figure out where you're stuck. So coming back to how people get budgets, people got budgets for Agile transformation, then they got budgets for DevOps transformation and for cloud transformation. And you can't sell DevOps twice, so you're bored. 

ADRIANA: [laughs]

SASHA: So you need a new transformation, so now there's microservices. After microservices, now there's going to be platform engineer, I don't know. So basically, there is this inherent drive to pick up the new, shiny thing. And a lot of times, it's well-intentioned. People are just trying to get rid of technical debt, and that's the only way they can pitch it. It's like, oh, we're going to rewrite everything, so we're on microservices. It's like, no.

ANA: [laughs]

SASHA: You actually just need to rewrite everything that's 20 years old, and you can't maintain it anymore. You don't need microservices. Your monolith is fine. It's an application that serves 100 people. You don't actually need Kubernetes for it. [laughs] It's fine.

ANA: And you also have those companies, like you said, apart from shiny things, the culture is about reinforcing those that build new things. So I've been at places that the more microservices you build, the more likely you are to get promoted. But it didn't necessarily have to do anything with the business impact.

SASHA: This is, to me, one of the dangers of SRE practice because SRE practice said toil is bad, so your general maintenance work, your work that you're doing day to day to keep the lights on and stuff. It's all bad. We want to automate it away, and we want to stop doing it. Or we want to minimize it. And then Google said 50% toil and 50% development of automating things and making things better. But I've seen companies that are like, oh, we are 20%, toil. Like, really? Do you think you can get away with 20% operations work? 

And it is a problem on a couple of dimensions because one is I just don't have enough time. Because I can't automate a thing if I don't know how this thing works. If I don't know what breaks, or what's brittle, or what's not functioning correctly, I can't write code to automate it because we just don't have the context. So you need that toil to learn what your system does and when it's not working to be able to make things better. 

And the second thing is what you said is like, if I'm only rewarding code, so if to get promoted I need to write new features, I have a horrible job. If I do toil, work like I do incident management 90% of my time, and this is what is actually making my company reliable, and then I don't get promoted because I didn't write code.

ANA: Yeah. And we still see that a lot in the industry, even through a lot of these transformations that we've gone through. So I'm curious to see how is it that, as an industry, we could start approaching it a lot more head-on? I know some folks who are tracking more of those metrics of deployment productivity of employees. And I don't necessarily fall into the box that you should do that and see it progress. But I also haven't come up with another alternative.

SASHA: Incentives are always dangerous. Designing your incentives is complicated. Honestly, there are a lot of companies that don't have any metrics at all, so people get promoted randomly or never. It's like, for smaller companies, there's a lot of my manager likes me kind of thing and whatever. And then you have big companies that are like, oh, you need to write about [chuckles] how your impact was good for the organization. 

ADRIANA: [laughs] So true.

SASHA: And it's just disheartening. Why is half my job proving that I'm valuable to you? 

ANA: [laughs]

SASHA: That shouldn't be the case. It sucks. So one of the things that we talk about at Ergonautic is you need a system of metrics. You don't need one metric, and you don't need a list of metrics that are disconnected. But you need to make sure that your metrics are kind of balancing each other. I'm not saying Microsoft was perfect, but at one point, I discovered that it was all about, like, oh, I implemented this new feature because people wanted to get promoted. And so it led to every department has their own testing tool because everybody wanted to get credit for implementing the new testing tool.

They don't get credit for using the testing tool. They get credit for implementing the testing tool, so that's the incentive. So they tried to change it to like the three-part thing, so it's like your contribution as an individual, your contribution to your team and how did you help your team and the organization, and your contribution to leveraging your team and the organization. So kind of trying to remove this lone wolf, I'm going to solve all the problems by myself. And, again, this is never perfect, but it's maybe a little bit better than incentivizing every person to write a pamphlet about how they're solely responsible for all the reliability at Google or something. [laughs]

ANA: I mean, it goes against hero culture. 

ADRIANA: Yeah, it's true.

ANA: Like, we're aware of the on-call hero. But at the end of the day, dev teams also have heroes like that where it's just like, land grab and try to do as much as they can, that amount of push the most amount of code, do as many reviews as possible. And it was like, but were you actually being a teammate and using features and thinking about removing tech debt?

SASHA: Hog knowledge, right? So no one can do what I do because I'm the only one who understands how this shit works. And I'm just not going to share any of this information. I'm going to roll my eyes at everybody else and just say, "Oh, you just don't understand." 

ANA: [laughs]

SASHA: There's a lot of that, unfortunately, in tech in general just because they were kind of rewarded for being smart. And they're still like...I don't think it's pervasive, but there are still pockets of this 10X engineer, whatever. 

ANA: Yes.

SASHA: If you remember "The Phoenix Project," Brent is this guy who knows everything. And there, he's not toxic; he's just ended up with all the knowledge. He's a problem. Like, he's a problem because he's a bottleneck for the entire system. So you're a single point of failure, human.

ANA: Yeah. [chuckles]

ADRIANA: Yeah. And it's funny because I remember reading that book, and I'm like, oh my God, how do you know my life? [laughter] Like, not me as being the bottleneck, but I've met so many Brents in my life, and I'm like, this is wild.

ANA: [laughs] That was definitely a great read. 

SASHA: Yeah, I gave this talk years ago, and I said there was a single person of failure. I only gave it once. I never gave it again. And it's like, someone pinged me six months ago and was like, "Oh, this is a great talk. I related so much." I'm like, oh, nothing has changed. [laughter]

ANA: I'm glad you saw my content, but oh no. [laughter]

SASHA: No, it's cool. I mean, I'm glad you liked this talk six years later, but like, oh no. [laughter]

ANA: I mean, I think we're always going to be seeing some of those things bottle up again, or it's like the same type of behavior comes up. Like, we don't learn from our past learnings, just as humans, until we fall all the way down deep. And it's like, someone else really points out there's a lot of Brents at their organization, and they can do something about it. 

ADRIANA: I do feel like at least now we're, in general, on a kinder, gentler path where these things are, like, there's awareness being brought to a lot of these issues. I'm not saying that they're necessarily being solved. But I feel like the fact that we have conversations about these things to surface them, I feel like there's hope. You know what I mean?

SASHA: Yeah. I mean, it helps, at least when people are aware and when people have the language to talk about certain things. Like, if you don't have the words or the idea is not generally accepted, it's really hard to explain what the problem is. Once there's some awareness, you can at least be like, okay, if we have this problem, here's a book on how is this a problem.

ANA: [laughs]

ADRIANA: Even when we're talking about things like psychological safety and SRE burnout and stuff, these are things that were just kind of accepted back in the day. And now it's like, no, we're having conversations about well-being. So that makes me happy. Because certainly when I started my career, this idea of like, oh, you can't push back because you got to earn your place in the company. And now I feel like we work in a more, I guess, accepting tech society where we speak out, and sometimes things get action, sometimes they don't. But at least I feel like people are more emboldened.

SASHA: Yeah, maybe; I don't know. I don't feel like I see so much of a trend. It's more like I feel empowered to change the place of work. So either I try to change the organization, you know if you can't change your job. Like, there's one of two things, either I'm going to make things better, or I'm just going to leave. 

And then the difference between a good organization to work at and a bad one are so vast. You don't want to end up showing up to work every day hating your life, and psychological safety is a huge part of it. And I think early in a career; you just don't have as many choices. You feel like if you leave after a year, that's going to look bad on your resume and this and that, whatever. And so I think that's some tenure freedom that comes with experience.

ADRIANA: Yeah, yeah, that's true. That's true. 

ANA: [laughs]

ADRIANA: I think the more established you get in your career, you just give less fucks [laughter] and care more about your own mental health and what you'll tolerate.

ANA: Even just being able to recognize what's terrible behavior. Like, this is not supposed to be long work hours for this long period of time, not just like here and there for finishing up a project. Those things end up catching on.

ADRIANA: Yeah, that's true.

SASHA: I mean, also in your 20s, sometimes you don't mind working a lot doing this kind of stuff. And then you kind of get older, and it's either you get more responsibilities in other ways, or you're just like, no.

ANA: [laughs]

ADRIANA: You're just like, no, I'm done.

SASHA: You're just like, this is the boundary.

[laughter]

ANA: The wall goes up here. This is my schedule. If you don't like it, I'm leaving the organization. [laughs]

ADRIANA: Totally.

ANA: I wanted to go back to some of the stuff that we were kind of talking about earlier, that we were joking around about all these buzzwords that kind of have come out of, like the DevOps movement, the SRE movement, and now the platform movement. With these buzzwords, what are some of those most cringe-worthy terms, in your opinion, that you keep seeing?

SASHA: I don't know about cringe-worthy. My brain isn't serving me anything specific. But I think we talk a lot about this continuous buzzword creation, so I'll give you an example. A lot of people think they do Agile, and they actually don't do Agile. But you can't come in and be like, "I'm going to teach you Agile," because they're like, "No, no, we're already doing this." "How are you doing it?"

ANA: [laughs]

SASHA: "Well, we have standup every day." [laughter] Okay. And the thing is there's no perfect methodology. And it depends what kind of organization you have and the size and complexity, and you kind of need to be able to adopt the practices to what matters to you. But also, I always say there's no perfect way to develop software, but there are bad ways. Like, certain things you're like, oh yeah, you should not do that. [laughs]

And then there's no...I don't believe in there's a gospel, and you should only do one thing only one way. But there are always ways to improve. So if you're familiar with the Theory of Constraints, you have these bottlenecks in your system, and usually, improving the flow through the bottleneck improves your system. But also, your bottlenecks sometimes shift. And so we had a bottleneck here, and now this has been relieved, so now you have another bottleneck there. And there's always something that can be better. I think the problem with buzzwords is this continuous renaming. So it's like, oh, we're SRE now. And again, people say, oh, whatever SRE transformation failed. And are you doing SRE? 

ANA: [laughs]

SASHA: Show me what you're doing. Have you tried this? Have you tried reading the book? [laughs]

ANA: Even when you read the book, they want to implement exactly how Google implemented it. And it's like, you're not even close to the size of Google, or you don't have that talent. You need to be more scrappy or just think differently.

SASHA: Well, so the other thing is the Google SRE was written by a particular subset of people at a particular point in time. It doesn't mean that all of Google does exactly this thing all of the time. And the other thing is, yes, the complexity and size and number of engineers, but the other thing that a lot of people don't have is this opinionated platform. Google actually says, "We're going to run the platform for you, and you're going to use it."

And I see a lot of organizations start working on a platform, but then it doesn't get adopted or something like that. Or the developers are allowed, for instance, to develop in any language they want. It's like, so do you think that developers at Google are allowed to develop in any language they decided? That's not the case. 

ANA: [laughs]

SASHA: Like, the operations burden increases exponentially if you keep adopting new technologies, and it's kind of like the wild west of what could potentially be happening out there. And I'm not advocating necessarily for, hey, limit everything, but there's this concept of enabling constraints. 

ADRIANA: Can we dig a little bit into starting your own business? I think you alluded to it a bit as to what motivated you. But how did it come about? How has the experience been?

SASHA: So it's interesting; I've been thinking about starting a startup for like a decade. I've never had a concrete idea of I want to do this or that. It just was an idea that I want to start my own company. I want to try how it feels like. There was a combination of things. I actually ended up working for bigger and bigger companies, and I'm like, oh gosh. [laughter] And it's hard to go try something on your own when people keep offering you promotions and salary raises, and you're like, this is going pretty well. Why would I go try something new and potentially make no money? 

ANA: [laughs]

SASHA: And then I think the timing was just right. So I got to director level, and I was feeling like I don't want to be a VP at a big company. This is not the career path that's exciting to me, at least not right now. Maybe sometime down the road, but not right now. And the farther you get up the management chain, the more it's about politics and less about, like, I don't want to dismiss it. Playing political games is kind of fun too. It's got its own kind of excitement, but it just kind of wasn't what I wanted. And then my co-founders just came around, and they were like, "Hey, we're starting this thing." And I was like, "I want in." And it's like, "Are you sure?" 

ANA: [laughs]

SASHA: "Yes, please. Can I please join you? This sounds amazing." So this is an adventure. Like, for months, I was like, I don't know if this is going to go terribly wrong or if it's going to be terribly right. It was kind of trepidation. It's so new. I always had a paycheck. I didn't have to worry about how do I do 401K? I don't know.

ANA: [laughs]

SASHA: How do I do my taxes? Everything is so, like, I'm doing research on the internet. How do you do this, and how do you do that? And it's like, in addition to you're trying to actually work on a thing. So it's like a totally new world. But it's been really exciting because we're building our own thing. We're defining our strategy, and we're defining what we're delivering and how we want to grow, and how we want to position ourselves, a very new world for me. I worked for a tiny startup before, but I've never owned one. So it's like a whole new world of possibilities here.

ANA: [laughs] That's so exciting. You get to put out this little project, feed it, and see it grow into this hopefully bigger organization that you get to be proud of.

SASHA: Yeah, yeah. And to me, I define success like if I learned a lot of new things and made some money; this is already a success. But if we actually get to do what we want, which is to grow the organization, scale it out, and hire a bunch of people, be really good at the type of consulting that we're delivering so we can actually help a lot of folks out there that will be just fantastic.

ANA: [laughs]

ADRIANA: That's awesome. And I do feel like you guys have a really cool little niche that you've hit up because there are so many organizations. I've worked for many large organizations, and it's like, yeah, they'll do their Agile transformation, their DevOps transformation, and then at some point, it tends to stall. So coming in and rescuing or righting the course, I think, is really valuable because we know the true power of what these kinds of transformations can give us. So giving those organizations the opportunity to really leverage those superpowers I think is awesome.

SASHA: Yeah. And I think there's kind of a couple of things that are different in being an external consultant, so one is I've sort of consulted in various ways for different organizations, but I always had, like, I have to sell the software, or I have to sell this thing. I have to do more of this or that, and you're beholden to a vendor in some way. So like, yes, you're giving the best advice you can give, but you cannot say, "Oh, you know what? You don't really need Kubernetes," because your freaking company sells Kubernetes. [laughter] 

And I'm completely free. I can recommend you whatever software I want or no software at all. I can say like, "You should start doing this. You should stop doing this." There are plenty of customers out there. We don't have an ulterior motive of like, oh, we want to sell you more consulting because it's not this kind of business. So it's a very freeing experience so far. And it's really really interesting to be able to talk to people really honestly about their experiences. 

And then the other thing that's really interesting is I just get to sort of come in from the outside. So when you're internal to an organization, for instance, I cannot come to someone in my own company and be like, "You're not actually doing SRE. You suck. Just stop it. This is horrible."

ADRIANA: [laughs]

SASHA: Because it's like, people are doing the best they freaking can. And I cannot come at someone and be like, you know what? You should really hire someone's prior SRE experience instead of trying to onboard based on the internet.

ANA: [laughs]

SASHA: Because sometimes it is. Andrew has this metaphor of learning a new language. You cannot learn a new language without having a person who speaks that language. You need that ability, that expertise, and feedback. In theory, maybe you could learn the Latin language without any...but this is completely different. If I can be immersed in a situation where people are actually skilled at a thing I'm trying to learn, my learning experience is so much faster and so much better. So that kind of thing, like, sometimes you need to bring someone in who's going to be able to set the tone for the team to be able to do something. 

So my point is, you know, I don't know if this is the best phrasing in the world, but the ability to say words and, of course, you kind of have to moderate. And you have to be able to message in a way that people can kind of relate to instead of being like, oh, everything sucks. You have to stop. But at least I'm not beholden to, like, oh, if I yell at these people, that's going to reduce my power in the organization. And then tomorrow, they're going to backstab me or something like that.

ANA: [laughs] It does make a difference. And, I mean, a lot of it too is that coming from someone who's done it before or teaching them by practicing as well, like, going hand in hand and guiding them and walking with them.

SASHA: Yeah, expertise matters a lot, and tools matter too. But it's the right tool for the job, not the shiniest new tool that you just saw.

ANA: [laughs]

ADRIANA: Totally. And as a quick follow-up, when you're coming in and basically telling teams like, "Hey, you're not doing this the right way," how do you approach it without them getting offended basically so that they're more open to the change? You're hired by a company, but it doesn't necessarily mean that the teams that you're working with are welcoming you with open arms. So how do you approach that?

SASHA: So, for one, I think a lot of times, people are just happy to tell you. A lot of times, people know what the problem is, but they have no power to change things. So a lot of times, you just arrive, and people are like, "This sucks, and this is why," And they're already correct, but they just don't know how to fix it. And then the other option is you take them on a journey. So you expose them to ideas and kind of make it their ideas. Or you teach them gradually, like, hey, have you thought about this? And then that requires some continuous conversation. 

But then basically, by the end of it, they're like, it's not a report that's coming from me, and I said you should do blah, but it's more of a collaborative effort where we thought together with you and everybody kind of thinks that this intervention is a good idea. And then the other thing is so when we do the report, we kind of list a lot of interventions, and we kind of list, like, okay, we think if you do these three things, it's going to be really, really powerful. And then there are those small fixes in different teams that you could do that could improve their life and whatever. 

And so you can choose your adventure. If you don't want to do the highest recommended thing, but you think if you fix this little problem, it's going to be better, you can do that. But ideally, you know, we haven't been in business for too long. So I can't tell you how many of the reports are like; I can see that people are actually taking it to heart. But I think the goal is to actually have folks implement it and have their lives be better. 

So that's the other thing is people are scared a lot of consulting companies because consulting companies come and do this big report that a lot of times is like, let's fire all these people, or let's replace all this software, you know, some kind of big things. No, you really should be better a little bit. You're not going to become an Olympic runner. But you could run a 5K after three months of training. That's kind of where we want to get you.

ANA: [laughs] That makes perfect sense. And that's a really good way to start thinking of putting little chunks of something new you're trying to do into your practice and having your team work towards it.

SASHA: Yeah, because the other thing we think about a lot is behavior change. Just mandating that people do something different isn't going to necessarily work. So how do you actually take people on this journey of learning a new skill and doing the thing? And ideally, there's automatically a reward in it. So I'm reducing friction. I'm reducing toil. I'm reducing the shitty part of your job, so you already want to do it because it's better for you.

ANA: That's very true. I mean, it's that incentive of life will be better after this, or once we finish this transformation portion, we get to have an off-site. We got to have perf reviews and bonuses or things like it. 

SASHA: Yep. 

ANA: As we're getting ready to wrap up the episode, is there anything that's exciting you right now about the work that you're doing?

SASHA: Yeah. So I think we're learning a lot of new things, and so one of my goals is...so right now, this is our expertise. So we have a few people who have done this for a decade or three. And they can come in, and they can give you advice because they've done this before. The thing that excites me the most is how do I write it down sufficiently, like what we do and how we do it that I can have a new person come in and we can teach people how to? So that's the exciting part is kind of getting knowledge out of our heads and onto paper. [laughs]

ANA: Spreading knowledge is definitely hard. And it's always nice to be able to have those conversations with people, like, how did you train other engineers to do this type of activity? Or how is it that you learn from incidents? How is it that you pass on deep knowledge of your architecture to someone else or even just the entire DevOps movement? [chuckles]

SASHA: Yep. 

ANA: Well, thank you very much, Sasha, for joining us in today's podcast. 

Don't forget to subscribe and give us a shout-out on all social media for oncallmemaybe. Be sure to check out our show notes on oncallmemaybe.com for additional resources and to connect with us and our guests. For On-Call Me Maybe, we're your hosts, Ana Margarita Medina...

ADRIANA: And Adriana Villela signing off with...

SASHA: Sasha Rosenbaum. Peace, love, and code.

ADRIANA: Whoo! 

ANA: [laughs]