Summer fun and podcast recap

How we won Atlassian’s Partner of the Year banner

As summer holidays roll in, Pinja and Darren take a step back from the usual deep dives into DevOps to reflect on the past six months of the DevOps Sauna. From endless AI debates and myth-busting episodes (“Is Jenkins dead?”) to accessibility, platform engineering at the edge, and even a full episode on sauna culture, they revisit the highlights, surprises, and fun along the way. They also share some of Eficode’s recent wins, welcome a new CEO, and admit to pulling off a very Finnish April Fools’ episode. Join us for a lighter, summer-ready conversation that celebrates both the serious and the silly moments from the podcast so far.

[Pinja] (0:02 - 0:12)

Whenever Finland has been mentioned in the news, we always say, let's go to the marketplace and celebrate, because we don't get talked about much in the international media.

[Darren] (0:14 - 0:22)

Welcome to the DevOps Sauna, the podcast where we deep dive into the world of DevOps, platform engineering, security, and more as we explore the future of development.

[Pinja] (0:22 - 0:32)

Join us as we dive into the heart of DevOps, one story at a time. Whether you're a seasoned practitioner or only starting your DevOps journey, we're happy to welcome you into the DevOps Sauna.

[Darren] (0:38 - 0:44)

Welcome back to the DevOps Sauna, where we're taking it a little bit easy today.

[Pinja] (0:44 - 1:01)

We are. It's not our usual topic here today, but since we're now tuning in for the summer, the summer vacation period is closing in on us. I think I'm starting mine in less than 24 hours, and Darren, you got a week or two to go, right?

[Darren] (1:01 - 1:03)

Yeah, a couple of weeks, and then I'm off.

[Pinja] (1:04 - 1:14)

Yeah, so let's take a look back at what has happened. If we think of the last six months, and the way that is set, that's when we started this podcast with this host composition.

[Darren] (1:14 - 1:34)

Yeah, we've had a bit of a changeover in the last six months. But yeah, we should maybe talk about how the year has been treating us, or the last six months have been treating us. And we should probably stick to things we're allowed to say on the podcast.

So maybe you should get started. How's the last six months been?

[Pinja] (1:34 - 2:07)

Oh, it's been different. Let's put it that way. Not only did I start in the podcast, this has been new for me.

I've been a guest in the podcast a couple of times before I joined as a host. But this has been a new role for me, looking at it from a more marketing perspective. And I've gotten to actually deep dive into really new subjects for myself.

I also started in a new role, personally, in October last year, so I became a team lead. So there's been a lot of learning, I would say, in the past six months. What about you, Darren?

[Darren] (2:07 - 2:29)

Yeah, a lot going on there. For me, it's been pretty much the same. I'm still a security consultant.

So I mostly tell people what they're doing wrong, and they ignore me. I don't know if I'd categorize it like that. But I guess now I have.

Yeah, I think it's been a good six months. We've had a lot of interesting topics to talk about on the podcast.

[Pinja] (2:29 - 2:36)

One thing that we must mention here, what we have covered is obviously AI. What else could we say?

[Darren] (2:37 - 3:09)

Did we cover anything else? It honestly feels like we've been talking about AI for the last 15 years with how much it comes up. I think I was actually at this Eurostar convention in Edinburgh, and the topic of AI exhaustion came up a lot, because everyone's attaching it to everything, because it's been a great sales tool.

But it's one of those things that will, in my opinion, will likely start fizzling out as people just become exhausted by the next AI offering, and quality will start becoming the key metric.

[Pinja] (3:09 - 3:38)

I believe so as well. And I'm just thinking about two years ago, when we had the DevOps conference, a global version online, and we had Martin Woodward from GitHub, and he said that we're going to see more change in the next five years, this was two years ago, than in the previous 40 years. And this is to say, it feels like we've been talking about AI now for 15.

Maybe that's the reason, one of those things, why it feels like the acceleration has been so great.

[Darren] (3:39 - 4:17)

Yeah. And let's be fair, he definitely wasn't wrong. Things have changed more in those two years than they did in the 40 before.

And AI is a huge thing, and it's here to stay. I don't think anyone's arguing about that. But people only have so much capacity for a topic.

Even those who are excited about AI, they only have so much they can put in. So I feel like we're starting to reach this plateau where we understand AI is here, and it's staying, but we maybe don't have to have it as the sole topic of marketing and discussion. And at least I'm getting to that point.

I hope companies are following suit.

[Pinja] (4:18 - 5:27)

We can see how the market is turning, because it's not, if we think of an actual consumer and a regular customer of a company, is it really AI that they want to buy? But rather, let's say, the service, the product behind it, might have been enhanced with AI. There might have been steps, let's say, complemented or substituted in the process of making the services with AI tools, agents, somehow in that sense.

But now companies seem to be a little bit more thinking about what is their take on it? What is the natural way for us to use it, rather than the number one, the first, I guess, was the, let's be afraid of it, and let's not touch it with a ten-foot stick, I guess, was the first reaction from many people. Then it was all AI, AI everywhere on everything.

Here's that sprinkle with AI for you. Here you go. I guess we're more, hopefully, moving on to that direction, what you just said.

It is here to stay. It is a natural part of our work life soon, hopefully. Every single time we have a news episode here with Darren, we don't have to start with, you know what?

AI happened. That's it, nothing else.

[Darren] (5:27 - 6:07)

That was a considerable theme. The frustrating thing for me, at least on this subject, is AI needs to extend so far beyond large language models, because that's all we're ever really talking about. I feel like there are cool AI applications in things like machine vision and actual world interaction.

We talked at one point about stealing one of the delivery robots from the S stores so that we could take it apart and figure out how it worked. These things work on basically programmatic AI. They're a similar concept to the Tesla, just on a smaller scale and less dangerous if they happen to drive past a school bus at full speed.

[Pinja] (6:07 - 6:44)

Yeah, that's true. Maybe going forward, we don't have to say that AI happened, and this is the latest on AI, but rather make the use cases and how the companies are actually using it, how development is using it. Because I really think that just the landscape of how many different kinds of applications are there at the moment that are related to development life cycle and AI in that.

There's a nice list from AINativeDev, and on AINativeDev.io, there's a landscape list of things. If one looks at it, you can see what kind of jungle it is actually out there when it comes to these tools.

[Darren] (6:45 - 6:48)

I haven't actually looked at this. We heard about this.

[Pinja] (6:48 - 6:49)

It's a funny grid.

[Darren] (6:49 - 6:51)

Okay, that is a wow.

[Pinja] (6:51 - 7:31)

We can put this perhaps link to the notes of this episode, so everyone can go. But if you start thinking about how many different kinds of process steps you can either supplement or complement with AI, and with all these kinds of tools, we're not just talking about GitHub's Copilot or GitLab's Duo, and these tools, Atlassian and Atlassian’s toolchain, for example, and let's say any agent that you can do. But it's every single step you're starting from, the actual product requirements, and you go all the way to go-to-market activities.

So everything in between those can now, in some way, be either complemented or supplemented with AI.

[Darren] (7:32 - 7:51)

Honestly, this is an impressive list. Even for someone who maybe suffers from AI exhaustion as much as I do, this is an interesting list. I've already got a new couple of tools I want to check out.

Okay, that's an interesting topic. It actually led to a fun few discussions. Was it back in February, where we had DeepSeek happening?

[Pinja] (7:52 - 8:18)

We had a colleague of ours, Henri Terho, one of our lead AI consultants, visit us in the podcast. So I think he coined one of the phrases that, the Chinese plagiarism machine has plagiarized the American plagiarism machine, I guess, was the gist of it. We were talking about the ethics of AI, and also Henri Terho went on stage to talk about AI in the conference, Future of Software, in March, in LLM specifically.

[Darren] (8:19 - 9:23)

Yep. Honestly, this has been a discussion point across many parts of the company, and we've had several people speaking on it. This idea that open AI would make statements that DeepSeek has ripped off its AI model while simultaneously being in various lawsuits for their stealing of copyrighted content is...

Actually, I talked about it myself on stage in Edinburgh last week, and there's actually this picture of me on LinkedIn that I assumed would get me into a lot of trouble that I don't think the leadership of Eficode have noticed yet. It's just me in front of my slide that says the words, AI is unethical, with no context. And I actually asked people from the stage to take a picture of me in front of that photo, just so I'd have to explain it.

But the idea is that the foundation of AI is so unethical that we, as the users, have to build safeguards around it to make sure what comes out of it is not unethical, that we have to be responsible.

[Pinja] (9:23 - 9:34)

And somebody said, oh, there's this bug bar with AI that is hallucinating, but isn't that a feature of the whole tool? Really?

[Darren] (9:35 - 9:40)

Hallucination is a requirement for creativity in AI. So yeah, it's 100% a feature.

[Pinja] (9:40 - 10:08)

If one really dives deep into what is an LLM, because we don't talk about enough of the fact that the tools that we're using are basically LLMs at the moment. And what do they do? They predict the next word.

And it's not that many days since I had to tell a close person in my life that you cannot trust everything that your AI predicts for you. Please make sure that you're actually dealing with a grain of salt. So we still have this learning curve to come through.

[Darren] (10:08 - 10:19)

Interestingly, for two people who always open up our news episodes by saying we're talking too much about AI, we did just talk for about 12 minutes about AI. So shall we move on to...

[Pinja] (10:19 - 10:50)

Yeah. What else happened? We covered the myth busting series of things.

Is X dead? We talked about Darren's favorite, Jenkins. Is Jenkins dead?

Then we continued with Agile, because we see all the clickbait articles and posts on LinkedIn. And then to kind of maybe not make our lives any easier, because this is called the DevOps sauna, and our company lives from DevOps. So we went through, is DevOps actually there?

[Darren] (10:50 - 11:14)

Yep. We actually covered a load of those. It's probably my fault.

It came after I binge watched several seasons of Mythbusters. So I guess that's how that happened. And if you're wondering if any of these things are in actual fact dead, I'm going to spoiler alert now.

No, they're not. They're all, all these things are clickbait, and we knew it beforehand. And I think you knew it as well, but I think it's good to address these things.

[Pinja] (11:14 - 11:30)

Just to be clear, unless of course you want to listen to those three episodes, go ahead and do that. We're not opposed to that in any way, but just to save you some time, and so that there is no cliffhanger, or that you don't get disappointed after those episodes, we're going to tell you beforehand, they're not dead.

[Darren] (11:31 - 11:43)

Yeah. One thing about that also is that if you've been listening to this podcast for a while, and have heard me say snarky and sarcastic things about Jenkins, it's worth listening to the Jenkins episode just to hear me say nice things about it.

[Pinja] (11:43 - 11:48)

Yeah, we gave Jenkins some love, so I think sometimes that's worth it.

[Darren] (11:48 - 12:02)

Yeah, it's one of those tools that's just like, it gets a lot of flack from a lot of people, myself included, but it's a good tool. It does what it needs to in the end. So go listen to that episode.

I at least had fun recording it.

[Pinja] (12:02 - 12:40)

It was a fun episode. All of those myth-busting episodes have been fun. Just kind of going, if we think of DevOps and Agile and our discussions and whether they're dead or not, we really had to go back to the roots and go to, what is the definition of Agile?

Where did it come from? Where did DevOps come from? Was it just something that Patrick Debois himself made up and now we believe in this cult?

Or is it dead because platform engineering has replaced it? And the same thing is Agile dead because we say that maybe Scrum and Scaled Agile Framework aren't the best ways of actually looking at business agility and the companies.

[Darren] (12:40 - 13:02)

Yep. It was interesting. At least the Agile episode taught me things about, well, Agile in general, I'm quite stubborn when it comes to these things.

So I actually learned something, which was a pleasant surprise. Shall we talk about May and the Accessibility Awareness Day? So that happens May 15th every year.

[Pinja] (13:02 - 13:36)

Every year, the International Accessibility Day comes to us and we had two guests from our UX team, our accessibility experts, Markus and Toni, and we talked about how actually accessibility is profitable for everybody and not just people who we might first consider maybe the recipients of, let's say, Accessibility Act, which is coming into effect, is it actually this month? No, in June, since the recording date is now in June. Big things are happening.

Compliance is required more than before. And that was maybe the topic that we don't cover enough within this podcast.

[Darren] (13:36 - 14:33)

Yeah, I think so. I think accessibility needs to be more of a direct discussion that happens more often than this once a year notification that, hey, accessibility is a thing, because it's something we all benefit from. There's accessibility tied into parts of our everyday lives.

I make the argument that I believe the mouse could be considered an accessibility device because before that, we were limited in using user interfaces. We could only use text-based prompts and they were slower and less responsive. So I think it's a good topic to talk about.

There's another topic that we, I don't actually think we've released yet. We've recorded it, but we have this series on platform engineering on the edge. And you may have just heard Pinja laugh slightly.

During the recording of that, we had to say platform engineering on the edge so many times. We couldn't come up with an acronym for it or anything. We said it over and over again.

It's lost all meaning for us now.

[Pinja] (14:34 - 15:15)

Yeah. I think it was the third episode that we recorded out of this, the series of three episodes here that Darren, you said that you'd never want to say this again. And I was so happy when we were making the notes for this episode and I realized I get to make Darren say this out loud at least one more time on air.

But we recorded three episodes on platform engineering on the edge. First of them being the 101. What is it? What do we do? What does it mean?

We then talked about the cases. We use cases of it from outer space to your refrigerator, actually. So really close to you. And then we talked about the security of things, obviously, because somebody in this podcast has written a thesis on this topic.

[Darren] (15:15 - 15:37)

Yeah. I certainly did. It's available somewhere on the internet.

But yeah, it's actually kind of interesting. We talked a lot about doing things that people aren't really, people might be starting to get into, like putting Kubernetes on edge-based systems and this kind of GitOps operations. So if you're interested in those, they're definitely worth a listen in my opinion.

[Pinja] (15:37 - 15:56)

Yeah. And every single month, we have put out this news episode where we've been talking about what is happening in the world of DevOps. I said many times that it has also been about what is happening in the field of AI, but we have not perhaps covered so much what has happened at Eficode.

[Darren] (15:56 - 16:07)

Yeah. I think it's important to have a small section. I think we've had some victories this year that maybe are worth mentioning, starting out with GitHub Partner of the Year.

[Pinja] (16:07 - 16:29)

That was in, I think, late autumn last year when this was announced. And this year in springtime, it was also announced at the Atlassian Team 25 event in Anaheim that we also won Atlassian Partner of the Year Award. So both major technologies from the DevOps perspective granted us a Partner of the Year Award.

[Darren] (16:30 - 16:41)

I think I'd like to focus more on personal achievements. We had Szillard Szell Cell winning a Tester of the Year Award, which was actually no surprise to anyone who's worked with Szillard. I think he kind of deserves it.

[Pinja] (16:41 - 17:16)

That is true. We also had Szillard here in the podcast. So if you want to go and listen in on that episode, that is available.

But also another personal achievement from our colleague Dinesh Singh Bisht. He has now earned the jacket for both Kubernetes and AWS. So he got the AWS Golden Jacket, I think, a couple of weeks ago.

Not that many people have both of these jackets, which are only available to associate little people who really showcase that they know this stuff. So this is a huge shout out to Dinesh.

[Darren] (17:16 - 17:20)

I think he might actually have the only AWS Golden Jacket in Finland. True.

[Pinja] (17:21 - 17:22)

That is true.

[Darren] (17:22 - 17:34)

Okay. I think the last bit of news we probably need to cover is the new CEO. We have a new CEO, Patrik Sallner.

I'd like to welcome him aboard and say thanks to our outgoing CEO, Ilari Nurmi.

[Pinja] (17:34 - 17:49)

Yeah. Thank you, Ilari, for everything you've done for the company. And we know that since the world of DevOps, especially in Finland, is so small, we know that we're going to run into you soon.

And then, of course, Patrik. Welcome.

[Darren] (17:49 - 18:19)

Okay. So if we're talking about podcast episodes, there are two that I don't think we're allowed to maybe draw much attention to. So I'm going anyway.

We're going to. April 1st, we recorded an entirely sensible podcast about the new AI method and maybe some of the language floating around regarding AI that I think a lot of people will agree doesn't actually make sense. So we decided to do an episode that didn't actually make sense.

[Pinja] (18:19 - 18:46)

No, this is something that, well, as Darren mentioned, this was April 1st. Read whatever you want from that information. But Eficode released a statement that we now have this new Suami method, spelled S-U-A-M-I method, and it is our agentic AI methodology.

And then we took our twist to it and we talked about the Torille agents.

[Darren] (18:46 - 19:13)

Yeah. I remember thinking up some weird acronym for that. I genuinely don't remember what it was, little history on that episode.

I don't think we should need to discuss the episode too much. You can go listen to it if you feel like listening to something weird. But basically our CMO Lapa, Lauri Palokangas, dragged us in and said, Hey, we're doing this joke.

You want to do something for it? And Pinja and I were just, yeah, okay. And then we just kind of made up some nonsense.

So.

[Pinja] (19:13 - 20:25)

It was completely sensible and just out of, I think like every kind of rule from improvisation applied to it. And whatever Darren said, I said, yes, and vice versa. And just for the ones who might not know what these acronyms mean in the Finnish culture.

So of course, Suomi is a way of pronouncing the Finnish name of Finland in a very, what's it, Tampere kind of way. There's a little dialect in there. Torille, this is a very Finnish thing.

So it means the market. And whenever Finland has been mentioned in the news, we always say, let's go to the marketplace and celebrate because we don't get talked about much in the actual media. So whenever something especially big happens, if Finland wins the world championships for ice hockey or something, I guess that's the only sport where we can do that at the moment.

But then for some reason, a bunch of people go to, especially in Helsinki, to our big marketplace and celebrate throughout the night and bathe themselves in the fountains and stuff like that. So that's where it's coming from. A little inside information.

[Darren] (20:25 - 20:33)

I think that's useful to know because it didn't dawn on me to explain this to people. I just realized we have, we probably have people listening who are not that familiar with Finland.

[Pinja] (20:34 - 20:38)

Yeah. We get really excited when somebody notices us outside of Finland. Yeah.

[Darren] (20:38 - 20:49)

On the subject of very Finnish things, we also at one point got a little bit bored with DevOps and decided to do an entire episode about the other part of our name, which is sauna.

[Pinja] (20:50 - 21:16)

Yeah. We really said that we're not fed up with DevOps. In the end, we, of course, said that, no, we're still going to stick to DevOps anyway.

We're not so bored with it yet. But I think this is most easily the most important episode of the year that we record. We were talking about the actual Finnish sauna culture and how it's something that is noteworthy.

And we even said that the BBC might've taken something out of our episode on saunas.

[Darren] (21:16 - 21:27)

Yeah. Weeks after we released that sauna episode, the BBC had some very similar messaging. So you can hear it here first.

We are clearly the superior news company.

[Pinja] (21:27 - 21:47)

Obviously. Obviously then there's no debate on that. If you kind of laid on listening to just an episode out of this list that we just mentioned and anything you consider like, Hey, maybe on my summer vacation or on my commute, I could give this a go.

This should be the one. Okay. Let's just be very clear about it.

[Darren] (21:47 - 21:55)

It's completely not what we usually talk about. So it will give you a good idea of what we would talk about if we had more free choice, but we have to cover DevOps.

[Pinja] (21:56 - 21:56)

Exactly.

[Darren] (21:57 - 22:13)

I think that's actually all we have for today. Just a quick, quick rundown of what we've done over the last six months and saying by, and hopefully by the time this episode is out, we might actually be back with the backlog. So what are you planning to do with your vacation time?

[Pinja] (22:14 - 22:23)

My plan is very typical for me and my family. We're traveling. I'm okay with saying this out loud because I think by the time this goes out, we're going to be back anyway.

But this is a...

[Darren] (22:23 - 22:26)

Well, Finnair keeps going on strike. So...

[Pinja] (22:26 - 22:42)

Yes. We're trying to maneuver flights around Finnair strikes. But I think we're, we're at least on the outbound at the moment, but that's, that's very typical for us whenever, when I close the lid of my laptop, we're going to be out to the airport anyway.

But what about you, Darren? What's your plan?

[Darren] (22:42 - 22:58)

Uh, not much. Probably spend some time out on the water, probably spend some time playing video games. It's...

Nice. Kind of typical. But yeah, we'll be back to recording, I guess sometime in August when hopefully this will have been released, but we'll see.

[Pinja] (22:59 - 23:23)

We, I think we, were trying to do the math that we would cover the whole summer vacation period without any gaps, but don't come at us if this didn't work out and we, we miscounted how many weeks we're going to be away. But thank you to our listeners and don't worry, we'll be back with sometimes more sensible than today. Sometimes not so much then we had to wait to cover.

[Darren] (23:24 - 23:28)

So... Hopefully not less sensible, but maybe a similar amount of sensible.

[Pinja] (23:28 - 23:37)

Similar amounts, but the, uh, let's just say the, the April 1st episode maybe was a pinnacle of nonsensical, nonsensical things that we, we have to cover.

[Darren] (23:38 - 23:39)

It usually is.

[Pinja] (23:39 - 23:39)

It is.

[Darren] (23:39 - 23:44)

All right. But yeah, thank you everyone for listening. Thank you, Pinja, for joining me again.

[Pinja] (23:44 - 23:45)

Thank you, Darren. It's been fun.

[Darren] (23:45 - 23:53)

And we hope you'll join us next time in the song. We'll now tell you a little bit about who we are.

[Pinja] (23:53 - 23:58)

I'm Pinja Kujala. I specialize in actual and portfolio management topics at Eficode.