Advice to the Young Minds

I decided to answer some of the questions I frequently get asked, here, in a single post.

  1. "Hello, here is my profile and I admire your work in so and so. I was wondering if you are taking up any interns this summer."

Hello and thank you for reaching out but I am really sorry I ain't taking any interns. Thanks again and best wishes.

2. "I am planning to apply to Masters program in Applicable Mathematics at London School of Economics. I was wondering if I can land into a successful machine learning/data science job after that?"

No, please don't. I beg of you. I owe a lot to the excellent professors I met during my time there who became my biggest supporters and mentors but you could meet them without enrolling at the school.

A degree doesn't guarantee you a job anywhere these days, without the skill set. Please instead spend your money at a good bootcamp, get a job and learn at work. After a few years, when you have more clarity on the missing gaps in your knowledge then go back. Do not do a degree for the sake of it.

Schools are setups for making workers and mass producing people who lack the skill sets required in the real world. Do not get trapped into it unless you know how to squeeze every penny out of this "evil machine" that traps the young minds.

Want to get a mentor for a career in ML/DS?

Listen to Jeremy Howard. As in, really, listen to him. Join and finish the courses. Do not look for the next shiny superstar and next online course or book. Jeremy is the most honest, selfless, straightforward guy who is deeply committed to his students' success. I stand by that a one hundred percent, if you actually care about my opinion.

If you don't know how to code, listen to Austen Allred and join Lambda School.

3. "Should I go for a graduate degree in Maths or Stats or Computer Science if I want to have a career in machine learning?"

I mean, it depends. There is no one right for everyone. I guess it's about first getting an awareness on your gaps in knowledge and then find the courses which fill that. Then, decide the degree and then the professors and finally the school.

Do not run after MIT or Stanford. Find out what you want and the people who can help you get there. But generally saying, if you come from a mathematics background, then pick a different major for your graduate degree like Computer Science or even Physics or maybe Psychology.

Most of the research is about getting inspired by different fields and applying that knowledge to yours. Now that's all about exposure, creativity and having the right tools in a supportive environment.

4. "I ventured into machine learning a year back and have been trying to understand and grab the quicksand of this field ever since, though with less satisfaction so far. Could you let me on what you did to pursue your inner cravings? Would love to talk to you about it and know more of your perspective on it. "

For the inner cravings, I guess you just have to sort of believe that every little action you take will compound eventually and keep on it. Regarding perspective, I don't know I really don't have any - the field is far too young yet and there's a lot to do. Stop overthinking and start doing a little every day and the mysteries would unravel themselves to you in due time.

5. "I have this idea of a startup and I was wondering if you might like to join in as a co-founder"

Thank you, but all the very best. I am more of a people-first person than an idea-first person and thus prefer working with people I already know and love as good friends.

6. "I watched your TEDx talk. Do you have any work or resource along the lines?"

No, I am really sorry. I did that TEDx talk when I was a young 21-year old "seemingly" well-off superstar in the bitcoin scene. That day, I sat in the audience feeling incredibly hollow, making my mind about quitting a job I hated (Though, I also met some nice people too and made lifelong friends working there). I resigned the same night and haven't looked back since.

Years later, I wish I had given the talk a few months later, which would instead then be about the key problems in HCI, since that is one of those things I am deeply curious to see shape up in the next few years.

Leave a Comment

Your email address will not be published. Required fields are marked *