Mouad Hadji

Hypothesis: the world's most valuable data is screen captures of outlier competent people going about their work. But very little of this data is recorded, let alone made publicly available. You should seriously consider recording all work you do, even if just for personal use.

I think there are lots of examples where generated software and UI will work on demand, in particular in a personal context when the stakes are low and it’s just a byproduct of a chat system’s response. But I would have different expectations in an enterprise context. The point of much of the software that we use in the enterprise is largely to just keep our processes on track and moving forward reliably. With high SLAs. These are some of the most foundational things to a business. How companies close their books at the end of a quarter, manage their client records, handle inventory, organize their contracts, pay their employees, build new products, and so on. For any of these workflows and data sets, you need a high degree of stable patterns in your workflows that you know will work every single time. Also, importantly, you’re paying the software vendor to think about the underlying process so you don’t have to, keep the software up-to-date with new regulations or industry trends, and so on. Most companies just simply don’t want to have to reinvent the wheel on HR, CRM, contract management, etc. And even when AI models could code these up on demand, they’ll do so just a literally differently every time, which would be quite chaotic for many of these workflows. And maintenance will be fully on you. Now, what I do agree with is that we will interact with our software in very different ways in the future. We will begin to prompt most of our software to get answers back, and we will have agents running in the background of most software doing work for us. When we interact with those agents, they will certainly frequently render interactions that are relevant to the workflows customized to us. But under the hood, there will still be deterministic systems for our core data and workflows, and that maintain the guardrails that AI agents adhere to. This is the future I’d bet on. Love this conversation,, and hope it keeps happening. Also happy for any arguments as to why I’m wrong.

you have to be blind at this point to not see the future. the era of traditional software, as a stable object with version numbers, roadmaps & feature sets, is pretty much gone. we’re prolly just slow to admit it. future systems will be fluid, ambient, entangled with context. e.g. you summon capabilities, no real “apps”. you speak, gesture, or hint but never have to learn a single interface again that’s not personalized for you. the end of software as we knew it is here, not near.

Job roles in 2027: We let LLMs focus on the “how”. We focus on the “why”. Programmer → Code Composer Before: Writing code line by line After: Designing logic & structuring systems Web Dev → Experience Designer Before: Building layouts and components After: Defining flow, behavior, and structure Data Scientist → Analytics Strategist Before: Cleaning data, structuring queries After: Framing questions, interpreting outcomes