That's the goal of representation learning, but we're not quite there yet. From a previous comment:

syllogism 264 days ago

Deep learning needs feature engineering too.

You still need to transform your context into a vector of boolean or real values, somehow. And that transform is going to encode assumptions about what information is relevant to the problem, and what's not.

Let's say you're trying to predict house prices. There's no end of geo-tagged data you might pull in. And if you have a cleverer idea than the next guy, your model will be more accurate. And, probably, if the next guy's at least competent, it'll be your feature ideas that set you apart.

In a linear model, you need to come up with a clever set of conjunction features, that balances bias and variance. You don't need to do that for a deep learning model, and that's a big advantage. But that's not the same as saying there's no feature engineering.

This is true, deep learning can make feature selection / engineering easier. That being said, a deep learning method can be over kill for a large number of problems that ML is used to solve. The amount of data needed for the training set and amount of computational power needed for the training set is often not available or a huge effort. I believe in keeping things simple if possible and spending a little more thought of feature construction. However, it isn't best for all problems.