Why yes, I studied Java at DURGASOFT with the certified instructor Mr. Nagoor Babu, how did you know?

Thank you very much gentlemen, you've helped me iron out my path for next year and a half or so. I was thinking of doing a python class after CS1 but think I will stick with Java until data structures.

>good results
In my experience, they don't get good results because they don't understand that Kaggle doesn't resemble reality.

Any problem which requires real science to solve has already been cornered by older software vendors and the remaining space (mostly in NLP and CV) is quickly being eaten up by Google and Amazon.

>Almost all models end up needing to be rewritten
Those days are almost over. Data Scientists are often getting software engineering questions in interviews and are asked about things like scalability and model serving because the goal of most companies who hire data scientists is to sell the product once everything is working.

>In my experience, they don't get good results because they don't understand that Kaggle doesn't resemble reality.

Big problem is a lot of them just don't know what to do when confronted with non-uniform data. They expect pretty datasets like kaggle but don't realize the abomination level of data they'll receive on the analytics end. There's a lot of pajeet tier retards along the way that manage to fuck everything up.The good data scientists prepare for that and have means of alleviating those problems.

>Any problem which requires real science to solve has already been cornered by older software vendors and the remaining space (mostly in NLP and CV) is quickly being eaten up by Google and Amazon.

I don't think the API endpoints are quite there yet. They're getting close and I even see some cool stuff through Azure these days. The solutions are pretty limited though, especially in NLP and CV like you said. We're getting better at packaging out-of-the-box ML solutions though.

For:

Outside of class what tools or skills should I develop to work with piles of shit?

If you're going down a Java path, then learn these in this order.

Core Java
Maven
Spring Boot
SQL (not just syntax but what primary keys are, what many-to-many/one relationships are, composite keys, et cetera)

I would say you should know these core things first. Getting anywhere near decent as these things takes a while. Try creating a CRUD service with React, Spring, and some DB like Mysql or something.

>a lot of them just don't know what to do when confronted with non-uniform data
>The good data scientists prepare for that and have means of alleviating those problems.
The only way through this nightmare is extensive domain knowledge, which is thankfully becoming a mandatory requirement for data scientists now.

I remember reading some thread on Reddit about data scientists almost killing a scientific software oriented company because they didn't understand anything but believed their hyperparameter tuned neural networks could solve anything without any knowledge of the domain.

Although this data boom has given me a bigger salary and more opportunities, I just hope sanity comes back soon.

I'm not going to say that I'm new and I need my hand held, but I'm kind of new and need my hand held. Where are the best places to practice SQL? Also I'm taking notes as most of what you said is Greek to me.

Attached: filterSalesmanTalksToPotHead.png (1280x720, 751K)

>best places for SQL
sqlzoo.net/

>other stuff
baeldung.com/spring-boot-start

Your priority should be to get junior position whether you are paid or not. There is only so much you can learn on your own without needing to solve a real problem.

There are many companies with legacy codebases written by Indians and Filipinos which is where you'll probably get your start, so just be aware that you need to crawl through some shit when you start to get a good job later.

In regards to data engineering, if you decide to go down that path. You need to know most data warehousing concepts and the big data stack (Hadoop, Spark and so on). It's not hard to learn but often interviews are algorithms and data structure based, so prioritize learning Java properly (you know the difference between a TreeMap and a HashMap and so on).

This is the information that is almost impossible to find among all the L2C/tranny sites.

Yeah, that's an ok starting point to learn about the graalVM

He actually looks good with the oil, much better than the fake tans usually look

Yeah but he still has that weird bronze shit on him. Even the spartans would look down on the buttfuckery that is modern body building.

Post durgasoft videos
youtube.com/watch?v=eTXd89t8ngI

Holy fucking shit! How in the fuck do people learn from this?