In late 2017, I hit a wall as a Data Science Leader. Everything in my life revolved around advanced analytics and machine learning, and yet, I had somehow made it an entire month without writing a single line of code or training a single model. Don’t get me wrong, I was working–and crazy hours at that–but I was no longer contributing technical products to the analytics community that had given me so much over the years.
There was another side effect. One that I couldn’t live with.
I had stopped learning. The team that I managed was growing, creating value and having fun doing it, but while I was busy building and supporting the organization, MLaaS tools like TensorFlow, Keras emerged. Automated parameter tuning libraries such as Studio.ML and Scikit-Optimize gained followers. Data and model pipelining technologies such as Airflow, Docker became required knowledge and finally, Data Science DevOps took off, supported by companies such as Domino and Data Science.com.
As a leader of my company’s corporate-level Analytics Team, I felt like I had lost touch. Years of education, experience and hard work had led me to…well nothing.
And so I realized after 6 years, thousands of miles, hundreds of sleepless nights– that I was I was done in that role. Not done in the “I need to take a vacation, do some yoga and re-center myself” kind of way. I was “being unemployed would be better” done. But…Since I live in San Francisco, the whole not working thing wasn’t an option, at least not if I wanted to keep a roof over my head.
So instead of retiring, I used my last bit of willpower and resolve to push myself through a series of grueling job interviews–all in the field of Data Science, Applied Machine Learning and/or Advanced Analytics. The technical interview process for Data Science Leader roles in the City was absolutely crazy–and perhaps the subject of a future blog post.
I needed to stack the deck, (Re) invest in myself, and make sure that I was ready and able to make an impact as a Data Science professional–both as a leader and as a practitioner.
So instead of retiring in my early 30s, I used my last bit of willpower and resolve to push myself through a series of grueling job interviews–all in the field of Data Science, Applied Machine Learning and/or Advanced Analytics. The technical interview process for Data Science Leader roles in the City was absolutely crazy–and perhaps the subject of a future blog post.
*Spoiler Alert*: I found a job–the right kind of job with most of the things that I wanted. …Now Cue the imposter syndrome, right on time! I instantly doubted that I was ready: mostly from the technical side of things.
The good news is that I was able to arrange the start date of my new gig to be 6 weeks from my last day at the old gig.
So, in the (rather large) shadow of one of my personal heroes Clare Corthell. I embarked on an open-source Data Science learning journey. If you’ve got the time and the desire, I highly recommend working through her full Open Source Data Science Masters. In my case, since I only have 6 weeks and I’ve already been working in the space for awhile (see my LinkedIn), I thought I’d try to design and accomplish a Data Science Sabbatical.
I know from my previous life as a Marine, that you can get a lot done in 6 weeks–especially if you are intentional about it. Personally, I’ve always wanted to develop true technical expertise in my field as opposed to just knowledge. A lot of folks who feel this way might just get another master’s degree or go for the PhD, but the opportunity cost of those options are prohibitive. Also I don’t want to get one because it might get boring.
There are TONS of Business Analytics, Data Science, Advanced Analytics, Applied Statistics certificates, bootcamps and other non-degree learning paths out there (usually at the cost of 10s of thousands of dollars), and I wouldn’t dare try to convince anyone of their merit (or lack thereof). But for the rest of us, the (mostly) neutral internet still provides.
The advantages of a DIY approach to learning are plentiful, but as a result some assembly is required. I’m sharing my story partially to provide proof to Future Justin (who BTW sounds awesome) that it actually occurred, but also in the hopes that someone else just might find it useful.
So without further ado…
The Data Science Sabbatical
Footnote: Nowadays, I’m sure most people have a pretty good idea of what “Data Scientists” do, but for the sake of completeness, here are some thoughts from some legends.
“A data scientist is that unique blend of skills that can both unlock the insights of data and tell a fantastic story via the data,” — DJ Patil
“A data scientist is someone who blends, math, algorithms, and an understanding of human behavior with the ability to hack systems together to get answers to interesting human questions from data,” — Hilary Mason
“By definition all scientists are data scientists. In my opinion, they are half hacker, half analyst, they use data to build products and find insights. It’s Columbus meet Columbo – starry eyed explorers and skeptical detectives.”