The toughest data science interview I ever had

I got bombarded for 45 minutes with theoretical questions:

🔸 Entropy
🔸 KL divergence, other divergences
🔸 Kolmogorov complexity
🔸 Jacobian and Hessian
🔸 Linear independence
🔸 Determinant

Continued 👇

🔸 Eigenvalues and Eigenvectors
🔸 SVD
🔸 The norm of a vector
🔸 Independent random variables
🔸 Expectation and variance
🔸 Central limit theorem

👇

🔸 Gradient descent and SGD
🔸 Other optimization methods
🔸 The dimension of gradient and hessian for a neural net with 1k params
🔸 What is SVM
🔸 Linear vs non-linear SVM
🔸 Quadratic optimization

👇

🔸 What to do when a neural net overfits
🔸 What is autoencoder
🔸 How to train an RNN
🔸 How decision trees work
🔸 Random forest and GBM
🔸 How to use random forest on data with 30k features
🔸 Favorite ML algorithm - tell about it in details

That was tough!

After the global success of the iPod, Steve Jobs wasn’t used to hearing the word “no.”

But when looking for the iPhone launch partner, that’s all he heard.

He had one last meeting with the only company who hadn't rejected him yet.

Here’s how he negotiated in that meeting 🧶👇

1) First, a bit of context:

Before the iPhone, the wireless phone industry could not have been more different than what it is today.

Wireless network providers (carriers) had pretty much all the control.

2) AT&T (then Cingular), Verizon, and others set the rules. They told phone makers how to spec the phones. They owned the distribution. They even controlled the phone’s software.

Phone manufacturers had little autonomy to do what they wanted.

Steve, obviously, had other plans.

3) He wouldn’t let anything compromise his vision of creating a device that combined an iPod, Phone, and Internet Communicator.

And, in classic Steve fashion, he wanted to be in control of it all.

4) Unwilling to compromise, Jobs knew that he needed to find a launch partner that would be willing to completely relinquish their control, in the hopes that his iPhone would be a massive success.

So he came up with his list of demands.

0/ [THREAD] Over the last 4 years, I’ve interviewed 100+ of the most successful Investors, Founders and Executives in the world.

Here are the 20 "must have" lessons that most stuck out to me.

Lessons on life, career advice, entrepreneurship and startups 👇👇👇

1/ Having a billion dollars is great, having a billion seconds is priceless.

Interesting thought experiment: If you had the opportunity to switch places with Warren Buffet, would you do it?

You would be a billionaire, but you would also be 90.

Time > Money.

2/ Always strive to simultaneously be overrated and underrated.

Contrary to popular belief, being overrated is good. It opens doors and gives you credibility.

But don’t let this go to your head. Stay hungry, humble and hardworking.

3/ Surround yourself with open minded people vs. close minded people

Open Minded People:
- Ask genuine questions
- Insatiably curious
- Listen attentively

Close Minded People
- Make statements instead of ask questions
- Focus on being right
- Speak more

4/ Get comfortable with being uncomfortable

If you get advice from enough people, the advice will cancel out. You can find 2 smart people to take opposite sides of the argument on virtually every topic.

Focus on developing your instincts and trusting your own intuition.

10 months ago we shut down Atrium after raising $75m in venture capital. Anyone hearing that knows I made tons of mistakes along the way. Someone asked me today what my biggest lessons learned were. Here they are:

Start with the mission
.

It is very hard to write the mission after the fact. You should start with a clear reason to exist and filter early hires for believers.

Start remote. SFBA is over for startups: the cost of housing and rent gives you much worse operating leverage. Many talented people choose jobs they want to be flexible re: location. Remote is better at this time in the market.

I wrote that before COVID, now it's extra true.

There is no skipping of the R&D phase of a company - if you try to skip this you miss the part where you are forced to develop something differentiated. Very hard to solve this with money.

Don’t build a services company. It’s more work to manage everyone and the reward isn’t there at the end of the day.

If you wanna level up - the fastest hack is to hangout with people who are ALREADY at that level.

The beauty of twitter is that you can hang with anyone.

The problem is ppl follow the wrong accounts. It's noisy as f*ck.

Here's 27 people you should follow (and why)👇

If you hang with smart, interesting, successful people - you will start to think & talk that way too.

We are monkeys that copy what we see. You can fight it, or use that to your advantage by surrounding yourself with people who you want to become more like.

Group 1: Smart, Rich, & Don't Give a F*ck (no filter)

@chamath - was early FB, has big balls, bet early on Bitcoin & SPACs

@rabois - kind of a dick but he's clearly smart

@balajis - this guy is from the future, and has come back in time to tell us what's going to happen

BTW - I'm recommending people who are:

* smart
* say interesting things
* successful (not just talkers)

Plenty of people are 2 of 3.

You need ALL 3 to make the list

Group 2: Business Buyers - they see the $$$s of 100s of businesses a year.

@awilkinson - built a $200M+ empire of internet businesses

@BrentBeshore - same as^, but non internet (unsexy) biz (eg. swimming pool repair, port-a-potty rentals)

@XavierHelgesen - he does both