Yep. All the no-code systems I've ever investigated are essentially expert systems, and if there's one constant over the history of AI boom and bust cycles, it's "statistical models absolutely demolish expert systems every time".

No-code platforms are really DSLs wrapped in a nice UI. No-code platforms that are more open and developer focused typically let you actually dump out the app as a big bundle of config/DSL (e.g. a custom JSON format).

Maybe using LLMs to generate DSL code will produce better (and more maintainable) results than fully-fledged languages?

I use n8n for managing ChatGPT's API and to connect it to other APIs like gmail. It helps abstract away the authentication side of APIs which can be very cumbersome to manage.

It also helped me better understand the chatgpt API. N8n allows you to code in raw javascript or python which has allowed me to branch out to pure python.

I would say that no-code low code has a place as a stepping stone, that can allow other invested parties besides developers to manage inter departmental flow and build a blueprint that can be redone with actual engineers in the preferred manner.

Seems to me that the demand would still be there, it’s just that AI would replace/augment whatever rules engine is mapping the UI into running code.

We’re in peak hype cycle for AI. You’re always doing to need cut and dry, deterministic processing of structured data. That’s what low-code is for in many cases.

You’ll hit a simple wall with AI, as with any tool, where you need domain knowledge and language to tell the thing what to do. The difference will probably be more SMEs who can tell the computer vs developers and product managers figuring out what the business wants.

Look at stable diffusion prompts. There’s an art to constructing the prompts - it’s just another form of programming.

IMO low code with AI could be the future, we also at DronaHQ are integrating AI capabilities in every possible way like having a dedicated Ask AI feature for database querying, writing JS code or connecting with AI tool APIs quickly and helping our users to decrease the building time as much as possible.

This is funny because "AI" has even less of a proven track record of efficacy in enabling non-coders to build applications than "no-code" and yet you're already giving it more credibility. It is just as likely that it doesn't work either.