Eventually, the Data Scientist title as we know it now will go away, and having specialized knowledge or skills will qualify you for jobs and differentiate you from the crowd. Depending on your interests or where you want to work, spending your precious time on one (or maybe two) of the following areas shows you have intention in your studies and will fit into your schedule. The focus areas I describe are what I think are the four broad areas that Data Science will split into. With that being said, read through the descriptions and ask yourself the questions in each section to see if it is a good fit.
Econometrics & Forecasting
Forecasting based on historic trends has long been an essential part of conducting business. Whether using seasonal averages or industry knowledge, predicting the future was always a use-case with numerous applications. Data Scientists have a good mix of talents that line them up with the request to fit forecasting models, but they have not always had the educational background to effectively communicate their results in terms of economics and finance.
To see if this area might be a good fit for you, ask yourself these questions:
- Do I like discussing current and future states of a market?
- Do I want to test hypotheses to show what drives sales or spending?
- Do I have any academic interest in economics?
- Do I like time series and forecasting problems?
- Do I enjoy interpreting models and the impact of features for other people?
If you answered "yes" to one or more of these questions, consider pursuing econometrics and finance. If you don’t have an educational background in economics, try reading an older version of a University textbook (save $). Your regression training will give you a good foundation to start from.
Computer Vision & Natural Language Processing (NLP)
From object detection to voice recognition, Data Scientists have been included as talent to support a companies needs in these areas. While both computer vision and NLP are interesting to just about everyone, proficiency in this area is specialized and requires some extra training and effort.
To see if this area might be a good fit for you, ask yourself:
- Do I want to know how computers interpret images?
- Am I interested in cutting edge model architecture for object detection?
- Do I want to know how various layer types (densely connected, embedded, recurring) work in a neural network?
- Do I want to read white papers on natural language processing?
- Am I committed to large portions of time doing research, continuing education and self-teaching?
If you answered "yes" to one or more of these questions, consider pursuing computer vision and natural language processing. The landscape of training for machine learning, computer vision and NLP is rich and diverse (books and online courses).
Engineering and MLOPs
As more fitted models start making their way to production, a knowledge gap in deploying and maintaining models emerges. A 100% accurate model that only lives on your machine is close to 0% useful. The ETL (extract, transform and load) and packaging of ML capabilities with requirements is a grey area for many current data scientists, and not always covered in training programs. Auto-ML capabilities are increasing the frequency that a useful model, designed to be consumed, will be thought of for production. In my opinion, this is an area of focus that will become the most lucrative in terms of jobs and compensation.
To see if this area might be a good fit for you, ask yourself these questions:
- Do I want to manage the deployment of models for other data scientists?
- Do I want expertise in Docker or Kubernetes?
- Are ETL and data storage interesting topics to me?
- Do I want to work closely with software engineers and web developers?
- Do I want to be an expert in the delivery and consumption of predictions?
If you answered "yes" to one or more of these questions, consider pursuing Mlops. I would start to learn data engineering and software engineering best practices.
Subject Matter Expertise and product management
Data scientists that are embedded in a certain product or business area eventually learn a lot about those areas. There are roles within companies that ask for Data Scientists, but are actually analysts with the capability of doing more with data. In order to increase your value in this space, being very good at managing the data science aspect of the business area and behaving like a product manager might help you. That way, you can do more than just execute tasks on a plan that may or not be data science tasks.
To see if this area might be a good fit for you, ask yourself:
- Do I want to know everything about a business process, product or service?
- Am I interested in owning a product with a machine learning or AI component?
- Do I want to be responsible for the resources needed to develop, build and deploy a product?
- Am I interested in ensuring advanced analytics and machine learning outputs align with business level KPIs and objectives?
If you answered "yes" to one or more of these questions, consider pursuing product management and more time learning about the products or processes at your current organization.
Conclusion
If you have the talent and time to do so, please enjoy being a master-of-all-trades. While some have the ability to do that, others may consider focusing their time in one of the focus areas above. I believe that doing so will prepare you for a job market that asks for Data Scientists with more specific skillsets.