I’ve seen some remarkable things while working at Orbital Insight over the past year. I’ve seen how a small group of motivated people spurred a Saudi Arabian sheik to respond to insinuations of OPEC cheating, helped Texas respond to one of the largest natural disasters in state history, and enabled the World Resources Institute to monitor deforestation across swaths of the globe. This tiny assemblage influenced these massive organizations with nothing more than training data, an AWS account, and loads of raw intellect. I saw first-hand how artificial intelligence and machine learning are tipping the balance of power away from entrenched institutions and toward ambitious skeptics pushing for a better way.

This realization caused me to throw myself into learning machine learning. I nagged my team every week to discuss the intricacies of our agricultural forecasting model. I implemented every fancy deep learning architecture I could wrap my head around in my free time. I joined the machine learning reading group to expose myself to the bleeding edge of research.

And I learned something about myself. Unlike studying chemical engineering in graduate school, reading machine learning papers was actually fun. Not only was the work interesting, but the authors were solving real problems. Facebook AI Research predicted the contents of photos using nothing more than hashtags. Google Brain labeled cell compartments without staining. Uber anticipated holiday ride volume around the globe with hardly any history. The list goes on with incredible diversity of thought and application.

However, the one thing these papers had in common was that the authors often claimed a Google or Facebook affiliation. After the fifth or sixth literature review where our team tried to intuit the intention of the author or fill in some missing piece of analysis, I decided that I didn’t want to watch this body of work develop from the sidelines. I wanted to work with the masters. I set a two-year goal for myself to join the ML team at Facebook or Google.

In pursuit of this goal, I enrolled in the Georgia Tech online CS master’s program, carved out time on my calendar to read papers beyond those we covered in reading group, and tried to solve even trivial prediction problems with various ML techniques to learn their benefits and drawbacks. My self-study was off to a fabulous start when the Facebook Applied Machine Learning team reached out to ask if I would be interested in chatting about joining the team. Interested? Of course. My two-year goal was materializing before my eyes in two months.

After a series of intense interviews with a supremely talented team, I received a call from Facebook inviting me to join (in Kanab, UT on the motorcycle trip if anyone is curious). I was ecstatic. I returned from the trip with a clear head, accepted the offer, and couldn’t be more excited to start this new chapter in my career. And yes, the self-study is still on. I’ll be taking my first course at Georgia Tech this fall.