Good to know if you are Transitioning to Data Science after PhD

Among people decided to transition from Academia to Industry I met two extremes: a group of people, who are convinced that PhD title is enough for them to be desired candidate in any company and another group, who are certain their work experience in Academia is useless. As usual, the truth is somewhere in the middle 😉 

Successful research is incredibly challenging. Making it through a Ph.D. (and beyond) requires significant mental and emotional growth. If you’re surviving the challenges of independent research, then you surely possess:

🏆 The ability to distill interesting things from large amounts of information

🏆 Structural thinking

🏆 Extreme attention to detail 

🏆 Strong planning and organization attitude

🏆 Serious mental fortitude to push through (and learn from) failure

Many of the skills you gain in graduate school align nicely with what makes for a strong data scientist. In both cases, you need to think critically about what questions to ask, how to get data, and how to extract meaningful insights from that data. And last, but not least, how to present your findings in a structural way. 

However for successful transition to industry you will have to unlearn some of the mentality inherited in fundamental research. 

🔴 The speed-accuracy tradeoff

The point of research is to uncover the truth no matter how long it takes. 

In your new role you will have to figure out what is “Good enough” for a given task.

🔴 Implementing the results of an analysis

It’s not enough to just generate knowledge. You also need to convince stakeholders to adopt your results, which calls on an entirely separate set of business skills. Often incorporating your results will also require some software engineering knowledge. 

🔴 Focusing on team above self

In academia your name is the most important asset and visibility is critical for your career success. It can feel disheartening to have your contributions invisible to anyone outside your company. Transitioning from independent work to collaborative coding with colleagues while adapting to best practices and product management can be challenging too.

🔴 Expanding skillset

During your Ph.D., you likely developed expert-level skills in performing certain kinds of analyses on certain kinds of data. Data science is like being a full-stack analyst: you need to be comfortable rotating between tasks and tools. Embrace not knowing everything, and get started on filling in the knowledge gaps.