May 13, 2026·4:191Making a case for PaaS in AI-Native World#futureofwork—4:19transcripttap a line to jump0:01Hello, hello, hello, hello. It is a brand new day. It's May 13th. Wednesday, May 13th. It's0:0849.47am and I'm recording from my desk. I just had a random thought. Well, it's not0:13random. I've been thinking about this for a while. But I think we need to start making0:18a case for platform as a service. So right now, if you're plugged into the stock markets0:25or the industry in general, there's a very strong sentiment against the SaaS companies0:31because the idea that the antropics or the open AIs are enabling more people to build0:38and build at such a scale that allows them replace their use of SaaS entirely with new product0:46or ideas that you're able to come up with using those LLMs is very likely. I think that that's0:54a viable threat to an extent. However, I think a line can be drawn at the point where we're1:01starting to look into platforms that enable workflows, platforms that are ingrained into1:07workflows optimization, customer management, employee management, service delivery, service1:14management, the ITSMs. I really think that those products and those companies still have1:20a very, very pertinent and a very, very critical role to play even in AI native, even in the AI1:26native world that people or the market aren't necessarily understanding right now.1:33Now, the first point I want to make is right now, it is incredibly expensive for any company1:40at whatever scale to invest to the max on Claude, for example. Recently, Uber talked about, yeah,1:50budget for the year and I don't know that companies are able to justify the spend based off of how1:57effective their products are, the revenue generation, ROI on their businesses. And so, like, there's still2:03like a huge industry vertical across multiple verticals that cannot afford to make those bets.2:10They still need deterministic workflows and that's where the platforms come in. So, the platforms still2:15help the companies run their business per usual, baking intelligence as much as possible, but they're2:21not necessarily maxing tokens or they're not necessarily overspending or going through any of these2:27things. And I think that this is a conversation that the market needs to start having more to really2:32understand the value that platform has in the AI native world. I personally think that in the AI native2:38world, platform companies are key collaborators or enablers of helping AI native scale, right? So,2:48what they're talking about scaling workflows and collaboration between humans and agents, for example,2:54platforms are at the perfect position to make this happen. This is my thought right now. I'm trying to3:01think more deeply about this and flesh it out a little bit more, but I wanted to drop a note so3:06that I can come back to this later and maybe do like a more in-depth thesis on why I think platform3:12is still the future, at least in the foreseeable near, like, I guess two to three years. I'll say3:18two to three years for sure. Two to three years. Like, if a company has over, if they have over a3:24thousand employees or they have over 10,000 customers, they still need, they still need the3:31workflow. I don't know that there's a company that will be able to bet on investing time, tech3:38infrastructure and the risk, security risks, the compliance risks, all of those risks that comes3:44with building a production grade solution for their own business. I don't know that a lot of customers,3:49a lot of companies can fully take that on while also delivering on their core value proposition3:56of whatever business it is that they are running. Mind you, not every business is a tech business.4:00Most businesses are service businesses or they sell something or they are retail or, you know,4:06they are a government organization, for example. You know, very, very nuanced. I think there's a4:12wider conversation to have here and I'm really looking forward to having more of this.