Posts

The Content of a Conversation

...is almost meaningless.

Have you ever been in a situation where you're so engrossed and deep into a conversation that one topic leads to another, and suddenly you're diverging into multiple things, and both of you speak together, and so, you stop the person you're conversing with and say, "sorry, you were saying?".

At this point I realised, what I was going to say barely matters. We were having so much fun just talking, that the actual subject didn't really matter. We'd carry on and move to another interesting topic where we could spend another couple hours talking. This is a bold statement to make, but the content of the conversation is never relevant. All that the conversation is facilitating is assessing and matching wavelengths. What you are talking about doesn't matter, as long as you are talkingand not just speaking - or even that, if at all.

Put yourself in various situations. Picture yourself meeting this Polish girl at work you immediately…

My Last Day at Tonbo Imaging | Just Another Day!

Image
There's a thing about final good byes.
My favourite movie over romanticises it,
how the last time you say goodbye at an airport is the most important,
that you have to look back, and look at her, look at life as it was, one last time.
Why is it that we want to make them special?
That we go as far as judging the entire journey
based on the last segment that got us there.
That we devalue everything,
if the last memory of it wasn't the best one.

I don't quite know what I was expecting on my last day at Tonbo. I had done all the handover, written all the documents I had to, talked to all the people I needed to a month before my last day. And somewhere in my mind, I was expecting (?) for something extra special to happen on the last day. Now you have to know, this was my first job, and I was pretty attached to the people and the work I was doing. I had spent a good 3 years at the place.
And so I suppose I was wishing for the skies above the office to clear up, for Jupiter, the…

A Letter To An Old Love

Maybe I'm over-romanticizing it,
Maybe I'm not.
Maybe we never know what we have until all we want is what we don't have
Maybe we knew and maybe we loved it

I don't miss you.
I miss the idea of you.
Maybe we can't ever go back to it
Maybe someday we might choose not to

I want our ground.
And I want you there.
I want it to rain a little,
and then for it to stop.

And for us to talk and not just speak.
Of things that mean nothing,
but mean everything to us.

Maybe they weren't of any importance to anyone.
Maybe we never cared.
I might throw it all away once I have it again.
Maybe you will do the same.

All I want is to think out loud.
And discover my thoughts all over again.
Without you trying to convince me of anything.
Without me trying to assert or force what I feel.

It's surprised me every time I think of it.
How you and I were the best at what we were.
And then we were strangers.
And how we still are.

Love is transient
and is something I don't understa…

The Human Operating System

Image
The Real-Time Human
Some of us need to be real-time-like systems - people who work in operations for example - making important decisions quickly, switching contexts and the job they're doing as soon as something of higher priority interrupts them. Take an air traffic controller at a busy airport for example. The person is bombarded with constant information, performs calculations, has to maintain composure under pressure and makes important decisions based on short-term memory in a fixed time.
The GPU-like Human
There are then GPU-like humans that can only do a very particular task, but do it very well. A very busy cashier for example - can swipe hundreds of items at a supreme pace, bill and move on to the next customer like slime rolling down the line - but as soon as one of the customers in line asks them a question, or speaks in a language they don't understand, the cashier crashes to a halt, almost as if they've malfunctioned and has a stress attack! (This happens, esp…

Don't Share, Just Do!

Image
You meet this friend you haven't met in a while for dinner one day, and you're at the table talking about what each of you is making of your lives. And so, naturally, you share with them your plans of this trip with your girlfriend to this small city in Germany where the beer is great, how you're working on writing this book comparing instincts of man vs animal, and how you're trying, albeit not too successfully, to work-out regularly for a healthier lifestyle.

Months pass, and you meet this friend again. And funnily enough, you're saying the same damn things again! You still haven't gone on that trip, you've barely worked on your book, and your body looks the same as it did the last time you met. You did, however, go for a couple of woodcutting workshops and ever since you've made plans to make this small birdhouse in your balcony. The plans are only in your mind, but you share them with your friend nonetheless.
Yeah. 6 months later, same story, same …

The Art of Problem Solving & Interviewing

Image
Relationships, kindergarten, school, university, and my job. All these institutions have tried teaching us one thing.
We've all been there. Solving problems. 

Whether it is resolving a conflict with your loved one, or your second-grade teacher, Ms. Carter, telling you to use your common sense, or a data structures question you're solving at university, or a real-life problem you're solving at your job. All you're doing in life is learning to solve problems. And the single thing each and every company in the world is doing is solving problems.

And so, it only fits that the best tech companies in the world base their interviews on this skill. Now, they need something a test for problem-solving that is
reliable,can scale to hundreds of thousands of applicants year on year, easy to gauge the interviewee's ability to solve problems, and communicate while doing so. Data structures and algorithms make for a really good method of gauging problem-solving abilities.
Having in…

Sublime is NOT a Code Editor!

Image
I was at Pycon Lt last week, and though there are so many good things I have to write about it, I couldn't help but rant this down. Half the speakers were using Sublime. And about 3/4ths of those attending whose screens I glanced at from afar were too. The remaining were using Vim. And that is just such a sad sight to see! Look, I get if you're one of the elites who've learnt Vim well and can boast doing everything there with great efficiency. But most don't learn. A good editor makes it impossible to not use its features. And have you seen what a real editor can do?
Here's why IntelliSense is amazing:
Amazing autocomplete that works
Python is dynamically typed and that makes it almost impossible for the editor to know what a variable is, or what it does. And yet, it's amazing how well IntelliSense does it. Jedi and youcompleteme for Vim are terrible.Inbuilt Debugger
Putting print statements to debug code is like knowing you have a stain on your clothes but instea…

How a Neural Network is like Munna Bhai & Classifying the Largest Number

Image
It took me a good week or so after going through the first 3 videos of the FastAI course for it to click in my head as to what was going on in a neural network. And now, the basics feel very foolish. On the side the past few weeks, I've been doing algorithmic and data structure problems for fun*, and I thought it might be fun to see if a neural network could solve some of these problems.

And so, I started out with trying to predict the largest continuous sum in an array.But to define it better, I'd make sure the array only had 100 numbers.But 10 numbers are easier for me to visually see. And so, the array would only have 10 numbers.Shortly after I realized that debugging this would take me too much time, and so I had to think of an easier problem.Okay, how about finding the largest number of a list.But, a regression problem is most likely going to give random numbers, and I'll have to use silly metrics likeRMSE which don't make too much sense in a question like this.An…

What 3 Weeks of Deep Learning Have Taught Me

Image
I've almost completed the FastAI course for deep learning, and here's a list of things I've learnt:
Deep learning is far from magic. So far I was convinced there was more to deep learning than just matrix multiplications and 11th-grade math. 3 weeks haven't shown me the signs.It's hard, or it feels like magic because humans struggle to visualize beyond 3D space.And so, for my first project, I made a simple classifier that predicts the maximum number from a list of 2 numbers. This can be visualized in 3D. More on why I did this in another blogpost, but in 3 lines here's what I learnt:
The truth is, you'll never really know if you should be happy with your output. That's because don't always know if you've reached the most optimal solution to a problem. It'll take your computer way too long to go check every value possible.It's very important that you use big long words that sound impressive otherwise normal people will think they can do it…

A Deep Learning Classifier for FIFA vs Real Football

Image
FIFA or Not? Try it out on www.rish.space/fifa-or-real


I took up the FastAI course a week ago to finally do some hands-on,state of the art, 10-lines-of-code-does-it-all deep learning. I took up many courses in college, joined Kaggle a good 4 years ago (sigh - the only problem I ever solved was the Titanic dataset and honestly didn't find it as fun as making an Android app), and now that I've made up my mind to do more DL, I want a taste of a regular deep learner engineer's everyday job. More on that later, this post is just about identifying whether an image is from FIFA or not.

Here's the why:
I've played FIFA all my life, starting with FIFA 2001 on my PlayStation2. By the time FIFA06 came out, my father used to say it was hard to make out from afar whether I was playing or watching football on TV. And so, I thought it'd be fun to see if a computer could do this.

What I did:
1. Ran a script to scrape URLs of the top 200 images on Google Images for FIFA, and ano…