In this post I explore some themes related to family and immigration through some discussion of my hometown, my family, and my experience growing up between Bulgaria and Silicon Valley over the past two decades.

I hope you enjoy!

I was born in a city called Ruse (Roo-Sé) in Bulgaria.

Ruse is a stately city cradled on the green banks of the Danube, Europe’s longest river. The city’s architecture was largely inspired by the other large European cities up the same river — Belgrade, Budapest, and Vienna. Due to its location, Ruse has long been a gateway to Europe for commerce, art, and culture.

Looking east across Ruse and the down the Danube River.

This city will always have a place in my heart. It’s where my parents met and got married, and where my maternal grandparents, great grandparents, and great great grandparents lived for generations.

If that weren’t enough, I spent my...

I haven’t written on here in quite some time — but, somehow, I got the inspiration to give writing about my last weekend a go in the style of a former post about hopping through Europe in 2017.

On this particular weekend, I went in with only two set plans — dinner on Saturday and dinner on Sunday, and the goal to fill in the rest of the time doing whatever popped up.

For a recount of which way the wind blew me over 60 hours in LA, read on!

Friday

Locations: San Francisco

Initially, the plan was to fly out on Friday night after work.

But, a friend was putting on a dumpling making night (to follow-up on our previous weekend’s sushi making) and when I did a cost-benefit analysis, I decided to stay in SF for Friday and bump my flight to the earliest possible one on Saturday morning. I wasn’t going to do much getting into LA after 21:00, anyway.

Exhibit 1.1. Dumplings (not ready to...

Over the past few months, I have been working at Lyft, helping spin up the Capacity team over here.

Recently, the work of our team caught some attention when Lyft’s IPO filing and the disclosure of our $300 million three-year AWS contract became the talk of the town. The top comment on the Hacker News post about Lyft’s S-1 discussed the contract, and luminaries like Sarah Guo and Benadict Evans tweeted their two cents on cloud infrastructure spending.

As a generation of cloud-first companies have scaled into huge businesses, managing the interface between those businesses and cloud service providers has become an important part of operating a software company, and the complexity has continued to increase.

In this post, I want to talk about why cloud is the right choice for Lyft, the levers that we are building in Capacity to understand and control the company’s cloud resource usage...

I spent twelve days in Europe, hopping around from town to town and thought I would catalogue some of my impressions and experiences. It was a pretty quick trip and I didn’t really get to know many peeps or even learn the names of 95% of the places and things I saw but it was also a really fun exercise in survival and a great way to spend an inter-co-op vacation.

Day 1 - Sofia, Bulgaria

• visited my awesome friends, Rada and Sneje. Only had a few short hours to catch up. They gave me salad and tea. Hooray!
• their apartment was really neat, with 180 degree views of the Vitosha mountain in Sofia and right by the budding tech district.
• grabbed some banitza and airyan with Rada on the way over to the airport bright and early in the morning.
• forgot my razor at Rada’s apartment. You will see my beard growing as the trip progresses; also, got fined 35 EUR for not checking into my flight three...

During my time at Facebook, I learned a whole lot about A/B testing. This post is a brain dump in which I go over the importance of A/B testing, the gist of the components and infrastructure necessary to support an A/B testing framework, and how A/B testing infrastructure is part of the broader data infrastructure that powers machine learning on the web.

A big boon when it comes to building web applications is the ability to iterate quickly. You build a version of your product that will do the trick. Afterwards, it is super easy to ship out incremental improvements, so you can make your product progressively better.

These changes can either be visible UI changes, such as changing the color of a button, or invisible backend changes relating to the algorithms that power whatever it is you are building.

The way we measure whether a product change is better is through something called an...

A neural network is a mathematical device that learns to approximate output based on training examples. It does so by modifying its internal state or variables based on the magnitude of the difference between the real value of each training example and the value that the network computed based on its present variables.

A deep neural network has at least two hidden layers. Finally, the term convolutional indicates that at least one of the layers applies convolution to the previous layer’s representation. We will talk more about convolution later.

TensorFlow is a library created by Google which allows programmers to define, train, and test neural networks.

The purpose of this post is to explain the function and implementation of a deep convolutional neural network in TensorFlow for classifying images from the popular MNIST data set. MNIST is a labeled data set of hand-written digits.

Going into my second internship, my goal was to understand what it takes to develop web applications that scale to millions of users.

No server can fulfill millions of concurrent requests. Thus, a scalable application must run on an arbitrary number of machines. Despite being distributed, we want our application to be addressable and function as if it were a unit. For example, although Facebook is hosted on many thousands of machines, the site feels as if it may be hosted on one mega-machine, being accessible in the same way (www.facebook.com) and having the same data.

The challenge of developing a scalable application is to create the impression of being a cohesive application – being accessible in the same way, having the same data – while running on multiple machines.

Let’s call a set of servers (from this point forward, a host) running an application a cluster. Each host in a...

Labor productivity has increased dramatically over the past several decades. In the most developed countries, only a fraction of the population now need be involved in the development and manufacture of goods; instead, people are employed performing services — managing financial assets, cutting hair, selling software, teaching schoolchildren, etc.

Everything points to the 21st century obviating the need for even more human labor by automating millions more jobs, such as those of drivers, doctors, lawyers, accountants, buyers, clerks, etc. Computers can simply do most routine tasks better; that trend is what is referred to as software eating the world.

Automation of labor is fundamentally good. People will need to do less work to create the same basket of goods, thereby improving society’s wellbeing. Any net growth in [productivity * labor] leads to increased welfare. When humans can do...

“There’s no innovation these days,” my colleague mentions in passing. Another echoes the sentiment. “Companies are doing whatever they can to squeeze customers now,” he continues.

I disagree. Two trends cause this misconception. People look to established players for innovation. And as the heuristics by which it works are unveiled, technology loses its magic; thus, working in the trenches of technology leaves us jaded to the broader innovations we create.

In reality, we are at the precipice of some amazing technological change. In this post, I suggest where innovation might take place in the coming two decades.

Development

The software ecosystem is great. Two generations of developers have streamlined the process of creating, collaborating on, testing, and deploying code. That paradigm is starting to spill beyond the world of bits and into the world of atoms. The process of open...