Platform Engineering 2.0

0:00 Hi friends, we have a special guest today. Luca, welcome to the show.

0:04 Thank you.

0:05 For people that don’t know, Luca is the founder of platform engineering. And can you tell us a little bit of the story? How

0:14 Yeah. Yeah. Yeah. So, we say more like core contributor to the platform engineering community, but yeah, like you know, founder, co-contributor, whatever. Um the well we were actually I was actually part of like a the product team at at Humanitech which was like a like a vendor and we initially started off the community as um sort of like our way of really like gathering intelligence around the product. We always kept the product and the community really separate which is why the the community really really started gaining steam completely independently. And in fact about two years ago we realized it really didn’t make sense to to to keep that that connection and we entirely separated them uh really like on an entity level as well. Um and so now I can focus fully on platform

1:00 engineering. That’s that’s that’s all I do. We have a bunch of different moving parts in the community. Um there’s platform con which is our annual gathering 40 to 50,000 people online two live days in New York and London about a thousand people each. Uh we also now announced actually the world tour um so we’re bringing platform con as an individual live day format to Paris to S. Paulo to Sydney and more coming soon uh towards the end of of of this year. Uh so lots of events we do house of cube um which is the the party here which is very fun. uh one of the core KPIs for for CubeCon, it’s always been how many people can we get hung over to day one of uh to to to of CubeCon. Uh we now brought it to reinvent, we’re bringing it to Google next in two weeks from now. So there’s a lot of event stuff that we

1:47 do. Um but I’m also really focused lately on our research arm. So we uh you know in the last six months we rolled out weave intelligence and that is the research arm of uh of the platform engineering community that’s really focused on a key macro trend that we’ve observing that we’ve been observing in the last six to 12 months which is really this idea that platform engineering is hitting this inflection point and it’s going beyond the traditional platform engineering sort of like infrastructure and devx focus and incorporating security observability obviously data and AI I those are the fastest growing segments in our community, right? Data engineers, AI people, security sRes. Um and so we’re really um we really wanted to give the

2:33 community and the market a more comprehensive way of looking at platform engineering and how platform engineering is becoming the operating model for the enterprise for the modern enterprise today. And that’s um and that’s really the mission of weave intelligence and what we’ve been focusing a lot on uh also here at CubeCon, right? uh we’ve we’ve been hosting a couple of panels and I did a couple of talks and and and really starting to present our research uh that we’re starting to publish.

2:59 Wow,

3:00 that was a lot.

3:01 No, thank you very much for doing this insight because I have a feeling that uh we are an inflection point.

3:08 What works in the past is not the same especially AI accelerated a lot.

3:14 Yeah.

3:14 And what how you keep up to date with all the changes that are happening?

3:19 Yeah. Yeah, exactly. Um, and and I think that’s one of the key themes that obviously we’ve seen here at the conference, right? Like everybody’s like AI AI. Um, and but I also think there’s many like the number of platform engineers is still growing a lot. Um, you know, most of the people that I got to meet in the last couple of days are say platform engineers right on their title. Um and um and I think that and so I think those two trends are actually very complimementaryary right um you know I think like we uh we were at this panel yesterday right and we’re kind of like looking we’re discussing how um you know in a lot of the trends that we’ve been discussing at CubeCon at many

4:06 industry events in the last five years or so have really been overshadowed in the last 18 plus months by AI when it comes to enterprise right like AI I is kind of like overtaken everything um except for platform engineering I think is one of the really key exceptions. Why? because AI is actually turbocharging the need for well-designed platforms, right? Um, increasingly enterprise organizations are realizing, hey, if I want to move from this impressive day zero, day one demo or prototype into something that’s actually production ready, that’s actually enterprisegrade, um, and that is that I can rely on for day 100 operations, not not day zero. Um, I need a well-designed platform. I need to uh take the same best practices

4:52 and the same principles that we’ve been talking about in platform engineering and apply them to agentic workflows, AI workloads, um and and really thinking about like how to expand and how do I redesign the golden paths accordingly. But like we were discussing yesterday, the same uh principles apply at the end of the day. Um and and I think that is that has been amazing because we’re seeing this huge interest in platform engineering um by everybody uh as as more and more companies realize this.

5:25 I definitely agree. Do you foresee that u in the future we have like a welldesigned architecture like vendor indep vendor independent that we can apply to any design because what it seems especially with bip code this new trend

5:42 something that is super easy to create like a PC right

5:45 or right

5:46 something that

5:47 start working

5:48 right

5:49 but the more we want to move toward production

5:53 it become a Frankenstein

5:54 yes Yes. Yes. And and I think it’s it’s uh it creates all sorts of new problems actually, right? Because like you can see from um so you know I was I was I was giving a talk yesterday around uh some of the research that we’re soon going to publish on wave intelligence. There’s going to be uh one of our first white papers on agentic coding in the enterprise. Um and there we sort of like outlined five different levels right of um sort of like agentic adoption and then what is the role of humans in that context. So I’m not going to go through all of them but you know level zero essentially is what you have right now like it’s only the human then level one is where you start bringing in the agents for example um and so you might have actually a speed up and and an

6:41 improvement locally right so in your own IDE in your in terms of your own code creation um but you know if you think about many developers doing that right and then they’re all feeding into the same CI pipelines in the same deployment uh flows um right? Like then very quickly you start actually clogging the system down the line uh when it comes to validation when it comes to like infrastructure creation and management and that’s already what’s happening today right and so to your point how do we actually take the same principles that we’ve been talking about in platform engineering and readapting them um to really serve this new uh this new setup and I think you know Gregor has I think always this helpful analogy of the paved road right of like hey if

7:27 you are without an internal developer platform even right so leaving agents uh aside for a second um if you think about the pre-platform engineering pre- internal developer platform setup you have um sort of developers that try to like need to figure out you know configurations infrastructure orchestrations a bunch of different things and so they can’t really move fast right so it’s kind of like driving on this like really you know messed up road with a lot of different potholes and so on and really the idea of the platform engineering initiative is to kind of like putting asphalt on this and and really like paving the road, right? The concept of paved road. Um so that now developers have clear guard rails and they can move faster without breaking things. That’s the IDP. So if you now think about how do we move from IDP to potentially ADP, right? So like

8:13 aentic development platforms. Um it’s really about taking the same concepts but like instead of like one lane highway now it’s like you know a million lane highways because obviously you want to you want to be able to support this you know 10x 100x increase in throughput uh that’s coming from agents.

8:32 Yeah this this is very interesting because I heard this um this analogy they say be how to become a 100x engineer.

8:43 Exactly. Because right now there’s this so-called corso introduced this yellow mode then everybody start going fast instead of approving every changes you auto approve. Yeah

8:55 and uh also Google release anti-gravity

8:58 right

8:58 so we are in this hyper accelerated world

9:01 and it seems that what was vibe code now is also specdriven generator code and there will be a new trend probably upcoming. Yeah.

9:10 So there’s a way to accelerate the development of application.

9:15 Yeah.

9:15 But we still need a well um nice architectural diagram.

9:20 Yes.

9:21 And following the best guideline.

9:23 Yeah.

9:24 Because what we saw is that it generate code but is it good code? Is it

9:28 right?

9:29 Yeah.

9:30 Is you know it’s kind of if it’s trained on the average Yes.

9:34 the quality of the code will be average.

9:36 Average at best. At best. Yeah. Absolutely. And to your point on the infrastructure. Yeah. like uh this is one of the things that we’re really focused on at Weave intelligence and broadly in the platform engineering community is bridging the gap between platform teams who are again traditionally focusing on infrastructure application developer experience SDLC right and on the other side data teams right and what we’re seeing is you know in top performing organizations already they’re kind of like hey you guys need to sit down at the same table and figure this out right like because we need to converge to one towards one unified platform from layer that is really serving both right um and and so that means obviously that then also the respective architecture right to your

10:22 points need to start converging um and I think there’s still some like pretty uh blatant uh question marks that I don’t think anybody really has answers for right um but I think to your point as an industry it’s important that we already really start talking about them and um and opening conversations about like, hey, this clearly is an applicable best practice that we can, you know, translate over here and the other way around, but then there’s still like a little bit of like a fuzzy hole over here and like and then, you know, have panels, have conversations, have research, start surveys. We are really ramping up the amount of surveys that we’re doing um because we just need to figure this out.

11:02 Yeah. This this is great. I think this is a great summary of uh the landscape that we are facing from now toward the future. Yeah.

11:11 It is nice because if you I’m speaking with vendors, everybody seems to have like a secret source.

11:16 Yeah. Yeah.

11:17 And by I’m really happy that you started this initiative platform engineering uh is the overall uh initiative worldwide initiative. Yeah. And how can people get involved and what are the option?

11:30 Yeah. Yeah. Absolutely. So I mean if people want to contribute to our research uh they can just go and weave intelligence and and they can get in touch with our analysts and um and start you know give us briefings especially for from leaders but also just generally like practitioners on the field in the enterprise uh love to hear from you. Um um you know and then otherwise there’s many many different ways of just learning from the community. We have platform conplatformcon.com is is free. you can sign up to the virtual event uh or come and join the live days uh that I was uh talking about at the beginning there uh there is a university so we have university.platformengineering.org or um that’s where you can access our certifications and trainerled courses, but we also are really ramping up the

12:18 amount of free courses um that we’re publishing um that are just open for everybody. And again that really reflect as much as possible as much as we can this idea of platform engineering becoming the operating model for the modern enterprise and specializing further into data platform engineering security platform engineering observability platform engineering and so many different nuances and and facets of this and that’s really what we’re trying to provide both with research at weave intelligence and with the uh learning material on the university and broadly on the events and the community. um so that we can really help people on their journey on their platform engineering journey and in their career.

12:58 Wow. I love this. I love I love all these type of resources and I’m sure that this community will be more happy to use it and contribute.

13:08 Yes.

13:08 How can people get in contact with you, your team and your organization in general?

13:12 Yeah, I mean uh we have a Slack as well. Uh so many of us are on the Slack platform engineering Slack. You can access it. You can sign up from platformeng engineering.org or from the website there’s a link as well but I never remember the link um and um and otherwise you know come find us at the events we have um uh you know LinkedIn obviously is always is always a good place there’s me Luca Galante on LinkedIn um and and and many of our teams are are there as well so yeah

13:40 cool thank you very much and have a fantastic 2026

13:43 yeah absolutely back at you and thank you for having me

13:47 cool bye