Malleable Software Will Eat the SaaS World

Summary

In the AI era, the winners won’t be the tools you adapt to — they’ll be the tools that adapt to you.Let's take Linear. It is a beautiful, well-designed, simple but inflexible tool with little room for AI to add value. AI thrives in messy, open-ended spaces where it can design, assemble, and adapt — but in Linear, the major design choices have already been made. At best, AI might shave a few seconds off repetitive tasks or auto-fill a few fields, but it can’t reinvent the core process, because the tool doesn’t allow it.(c) LinearLet's take Fibery. It is somewhat beautiful, quite well-designed, complex and flexible tool 😝. However, it is relatively hard to setup Fibery for your needs. LLMs turn complexity from a barrier into an advantage, collapsing weeks of setup into a few prompts. In a world where “how” disappears, the most adaptable tools will win.(c) Alex (from Fibery)Problem vs. SolutionThe biggest shift LLMs bring to malleable software is moving the focus from designing the solution to defining the problem.In the past, when you had a problem in mind (the what), you still had to figure out the how — which meant learning the tool, assembling components, and translating your needs into its language.Now, in many cases, LLMs can handle the "how" for you. You describe what you want in plain language, and the system works like a programmer or system analyst: breaking your problem into building blocks, mapping a flow to solve it, and creating the first version. You review the result, give feedback, and iterate. The entry barrier drops dramatically, and the loop from idea to working prototype becomes fast.Malleable software FTWHistorically, malleable software was a niche for tinkerers. It demanded time, patience, and a willingness to wrestle with complexity. Learning it took effort. Building with it was cognitively heavy.That’s why simple, vertical solutions thrived. You picked something popular, used it as-is, and avoided customization altogether. Linear, for example, ...