Can you help me understand, if libraries like pandas and numpy also applies to your comment? Or are they truely optimized and you’re just referring to the standard Python language?

Here’s the plan:

Run the following models:

- Speech-to-text - Text-to-text - Text-to-speech - Text-to-image - Image-to-text - Text-to-video - Video-to-text

Start by integrating third-party APIs, and later switch to open-source models.

Implement everything using your preferred backend language. After that, connect it to a frontend framework of your choice to create interactive interfaces.

You want use your own data? Put it in a database and connect it to your backend, and run these models on your database.

Once you’ve done this, you’ll have completed your full stack development training.

I think this is a great take. Those problems have traditionally been hard to solve in engineering and you can get pretty reliable solutions from just an api call.

From what point of view is it bad? Technical?

It's a bad idea in general. Users would be annoyed, AI generated code is unreliable, just hire human engineers and listen to customer feedback.

Those are features. What human business problem are you solving?

Don’t put the cart before the horse.

Yes, it's important to start with a clear problem to solve. The ideas are great if they solve a real problem in an innovative or more efficient way than current solutions.

Sure you get paid more, but it comes attached with a whole baggage of other things.

A better research should provide insight on the total compensation over the entire career, work-life balance, impact, satisfaction and so on.

Now let's see what is better.

Downvotes help people skip the noise and focus on the signal. Ain’t nothing wrong with that.

I think what he means is that many of the seminal work related to cognitive science were produced back then. For example, Chomsky, Minsky, John Searle, David Chalmers and many more.

Things still move during winter, just not as much.

What step do most find the most challenging?

For me, it's the first step. It seems once you get started, everything else tends to fall into place more smoothly.

I'm eager to know others' perspectives.

Elon had always advocated for a for-profit structure so that OpenAI can raise enough money to achieve its vision. In the email they are discussing how the for-profit structure could work. Nothing wrong with that especially when he was the GP.

I much rather talk to an AI bot than waiting on the line for a human for 50 minutes.