Ask HN: Who wants to create a GPT-first company?

Some years ago I thought it would be interesting to crowdsource company creation. After using GPT4 for software development, I thought why not create a GPT4-first company? Here is the company: automate the process of architectural drawings for homes. Currently, if you want house plans your options are to hire an architect, hire a drafter, or buy stock plans from one of a dozen websites that sell them. The problem is that the architect takes 6 months and costs quite a sum, the drafter is much less expensive but can't give you their attention since they are super busy, and the stock plans tend not to be quite what you are looking for and are an inconsistent hodgepodge collection from various design programs. We know there is a solid market for this. I don't think the competition is architects but rather the stock sites. (Architects will be needed to design libraries, schools, chic homes, and large glass skyscrapers.) The customer is someone who wants a plan but doesn't need the full services of an architect. There are a lot of jurisdictions, in the US anyway, that do not have any zoning requirements which further narrows down the initial target market. Why me? First, this is a huge pain that I've experienced. Like many fields, in architecture you see a wide variance of skill and it seems that it matters not whether you went to an Ivy but rather whether you care and are determined to stay up to date with building science and technology. Few of these people exist; good luck finding them and hiring them. If this software existed I'd absolutely have used it and paid serious money for the final plans. It would have saved me headaches, months of time, and money. Second, I'm a big fan of high performance homes (https://news.ycombinator.com/item?id=34041900 https://news.ycombinator.com/item?id=34408123) and think it'd be advantageous to have these elements as design features. I've been able to successfully implement proof of concept IBC2018 CAD designs in Python using GPT-4. Inductively, this could work! So by "GPT-first" company I mean that the team would explicitly use GPT or another LLM (IDEs and such being developed) as a tool on a consistent basis. There could be other uses of AI like CV for interpreting a customer's drawing or a customized LLM for the user to dictate what they want. Note: there are a lot of projects that are attempting to use GANs or stable diffusion for the purpose of design, and I think that direction is misguided. Interior design may be an interesting avenue to go eventually, but any design should be discrete and not a non-deterministic result of a trained model. The founding team should be a mix of ui/graphics designers, generalists, and cv/ml engineers. An architect or drafter will be important for evaluating results. Since there is a high performance home angle, it probably helps to want to have a positive effect on the world in terms of carbon emission. It'd be a fun experiment. Who is interested?