Advice for Data Scientists on Where to Work

Advice for Data Scientists on Where to Work

by Eric Colson, Brad Klingenberg, and Jeff Magnusson on March 31, 2015
from San Francisco, CA

It’s a good time to be a data scientist. If you have the skills, experience, curiosity and passion, there is a vast and receptive market of companies to choose from. Yet there is much to consider when evaluating a prospective firm as a place to apply your talents. Even veterans may not have had the opportunity to experience different organizations, stages of maturity, cultures, technologies, or domains. We are amalgamating our combined experience here to offer some advice – three things to look for in a company that could make it a great place to work.

Work for a Company that Leverages Data Science for its Strategic Differentiation

Companies employ various means of differentiation in order to gain a competitive advantage in the market.
Some differentiate themselves using price, striving to be the low-price leader. Others differentiate by product, providing an offering that is superior in some way. Still others differentiate by their processes – for example providing faster shipping.

A Data Scientist should look for a company that actually uses data science to set themselves apart from the competition. Note that data science may be supportive of lower prices, better products, and faster shipping, however, it is not typically the direct enabler of these differentiators. More commonly, the enablers are other things – economies of scale in the case of lower prices, patents or branding in the case of product, and automation technology in the case of faster shipping. Data science can directly enable a strategic differentiator if the company’s core competency depends on its data and analytic capabilities. When this happens, the company becomes supportive to data science instead of the other way around. It’s willing to invest in acquiring the top talent, building the necessary infrastructure, pioneering the latest algorithmic and computational techniques, and building incredible engineering products to manifest the data science.

“Good enough” is not a phrase that is uttered in the context of a strategic differentiator. Rather, the company and the data scientist have every incentive to push the envelope, to innovate further, and to take more risks. The company’s aspirations are squarely in-line with that of the data scientist’s. It’s an amazing intersection to be at – a place that gets you excited to wake up to every morning, a place that stretches you, a place that inspires you (and supports you) to be the best in the world at what you do.

Work for a Company with Great Data

In determining what will be a great company to work for, data-science-as-a-strategic-differentiator is a necessary criteria, but it is not sufficient. The company must also have world-class data to work with.

This starts with finding a company that really has data. Spotting the difference between data and aspirations of data can be especially important in evaluating early-stage companies. Ideally you’ll find a company that already has enough data to do interesting things. Almost all companies will generate more data as they grow, but if you join a company that already has data your potential for impact and fulfillment will be much higher.

Next look for data that is both interesting and that has explanatory power. One of the most important aspects of your daily life will be the extent to which you find the data you work with compelling. Interesting data should require your creativity to frame problems, test your intuition and push you to develop new algorithms and applications.
Explanatory power is just as important – great data enables great applications. There should be enough signal to support data science as a differentiating strength.

Finally, don’t fixate on big data. The rising prominence of the data scientist has coincided with the rise of Big Data, but they are not the same thing. Sheer scale does not necessarily make data interesting, nor is it necessarily required.
Look for data with high information density rather than high volume, and that supports applications you find interesting or surprising. This enables you to spend most of your mental energy on analysis and framing rather than on efficient data processing.

Work for a Company with Greenfield Opportunities

When evaluating opportunities, find a company that doesn’t have it all figured out yet. Nearly all companies that fit the criteria in the sections above will already have some applications in place where the work of data scientists is essential. Look for those companies that have a strong direction and strongly established data science teams, but have an array of problems they are solving for the first time.

Often the most exciting and impactful opportunities for data scientists at a company are not being actively pursued.
They probably have not even been conceived of yet. Work somewhere that encourages you to take risks, challenge basic assumptions, and imagine new possibilities.

Observing the relationship between engineering and data science teams is a quick way to determine if an organization adopts this mindset. Is engineering enthusiastic to partner with data science teams to experiment and integrate ideas back into the business? Is there an architecture in place that supports agile integration of new ideas and technologies?
In fact, in companies that embody this mindset most effectively, it is likely difficult to locate the boundary between data science and engineering teams.

A greenfield can be intimidating in its lack of structure, but the amount of creativity and freedom available to you as a data scientist is never greater than when you’re starting from scratch. The impact of putting something in place where nothing existed previously can be immeasurable. Look for chances to be involved in designing not just the math and science, but also the pipeline, the API, and the tech stack. Not only is creating something new often more challenging and rewarding, but there is no better opportunity for learning and growth than designing something from the ground up.

Incremental improvements have incremental impacts, but embrace the chance to operate on a greenfield. While it is extremely important to constantly iterate and improve on systems that already exist, the Version 1 of something new can fundamentally change the business.

Summary

Of course, there are other considerations: domain, the company’s brand, the specific technology in use, the culture, the people, and so forth. All of those are equally important. We call out the three above since they are less frequently talked about, yet fundamental to a data scientist’s growth, impact, and happiness. They are also less obvious. We learned these things from experience. At first glance, you would not expect to find these things in a women’s apparel company.
However, our very different business model places a huge emphasis on data science, enables some of the richest data in the world, and creates space for a whole new suite of innovative software.

Originally published in the Stitch Fix Tech Blog

You may also like...