3 ms·
I'm a technical founder working in a deep-tech startup. If I am completely honest, I think the recommendation of having a technical cofounder while useful for
by syrusakbary 3y ago
I'm a technical founder working in a deep-tech startup.
If I am completely honest, I think the recommendation of having a technical cofounder while useful for startups in the past might not be as useful for the startups of the future.
Why? I strongly believe AI will cause a paradigm shift, and I'll adventure stating that AI will make non-technical people be able to do more and more with less. Which will make the requirements for differentiation in the tech side even harder to accomplish (very few people reaching the god-tech realms), but at the same time it would make the bases reachable for almost anyone. Basically, a more polarized order where less people will have access to "technical founders" that can differentiate themselves enough outside of the AI realms.
I'd love to get more thoughts on this!
- WendyTheWillow 3y agoHaving used LLMs to solve tech problems, I would not trust the current state, or even GPT5, to reliably solve anything other than the most trivial of trivialities. Is the idea that in 2025, AWS will have an AI console where a non-technical founder types in what they want their site to do, and the LLM takes care of the rest? Register domains, build out the database/backend/auth/frontend, etc.? I really can't see it happening in 2025, or 2035, without some pretty big leaps forward, the kinds of leaps that may never be possible.
- danenania 3y ago"Is the idea that in 2025, AWS will have an AI console where a non-technical founder types in what they want their site to do, and the LLM takes care of the rest? Register domains, build out the database/backend/auth/frontend, etc.?" Even when this AI exists, the non-technical founder will ask it for the wrong things.
- TuringNYC 3y agoLOL? Why do we even need the business co-founder? Let the LLM create the idea also and keep 100% percent of the equity for the FAANGs that created the LLM
- vlod 3y ago>working in a deep-tech startup. I'm curious if you use it for non trivial stuff? I lot of people say it will take over, but I don't see many devs using it beyond simple stuff.
- syrusakbary 3y agoAI usage is trending up. I don't see any signs of deceleration in our company. Right now is mostly trivial, but as AI is able to get more context on the space, it will go up. I foresee AIs learning from the code of your repo, your language, your dependencies... and most importantly, your competitors (if they have open-source code of course). The more context, the better it will get, the faster you will iterate and the bigger you may get.
- manicennui 3y agoComments like this make it difficult to take the HN community seriously.
- syrusakbary 3y agoHow so? If you think I'm missing something important, I'd love to hear it!
- danenania 3y agoIn my experience, usually the problem with a non-technical person isn't that they can't implement their ideas. It's that even if they could, their ideas are unworkable for a hundred reasons that they can't begin to understand. So why would the outcome be any different with AI? Even if the AI is a super-intelligent system designer and tells them all the flaws with their thinking, they'll tell the AI to shut up and get to work just as they'd do with a software engineer. I actually think that for the foreseeable future, the opposite is more likely. A technical person will be required to get good technical results out of AI, whereas a lot of the key growth-related grunt-work that a good non-technical founder takes on will become easier and easier for a skilled engineer to automate.
- syrusakbary 3y agoInteresting. If I put it in other words, the argumentation is that AI will just propel the already existing qualities of the people. And because people are different (and the distribution of intellectual qualities remains constant as we grow), the AI will not make any difference on that point. I'd agree on that. However, let's imagine the scenario of two people have the similar qualities, with one main difference between them: one have worked longer on those than the other (but their "base" is the same). Do you think AI will propel them at the same speed? (or said in other way: where will be the biggest delta on speed between their pre-AI and post-AI versions?)
- danenania 3y agoI think the key factor is how difficult it is to specify what you want in detail, since this is still going to be necessary with AI. If you can't clearly express what the AI should do, you won't get good results. I'd say it would be much easier for a technical founder to write a detailed spec for the work that a typical non-technical founder does than the reverse.
- mejutoco 3y agoIMO the problem is defining oneself as non-technical. Anyone can learn any topic if willing and with necessary resources. Some people make it part of their identity that they wont learn certain skills. Tech is nowadays like literacy. Imagine a person saying I am not literate, instead of trying to learn to read and write, at least to a degree. I think it is a status problem. Some people think of management as higher status than operational roles, and they do not want to be seen doing the low-status tasks. While it can pay well, tech seems for the general population as respectable as a mechanic or a chef, but not as a lawyer or a doctor (generalizing too much to try to get the point accross) On another note, sql was supposedly intended to be used by "business people" but developers are instead the ones using it. In general, a business type wants to loosely specify the needs and get some body worry about the details. I am talking here about the typical archetypes. Of course the best business people are not afraid of digging into details and learning, but for the average ones I found some generalizations apply.
- esafak 3y agoWhy is the technical founder becoming obsoleted but not the business founder?