I’m amazed by how many transitioning into #datascience aren’t aware of the specialities that exist within the field. Careers within #datascience - A thread:
1/ These are the main careers within #datascience based on my experience and which will likely continue to evolve.
1. Product Analytics DS
2. Generalist DS
3. Specialist (NLP, Computer Vision)
4. ML Researcher
5. DataOps Engineer
6. DS Product Manager
7. Data Vizualizer/CC
1. Product Analytics DS
2. Generalist DS
3. Specialist (NLP, Computer Vision)
4. ML Researcher
5. DataOps Engineer
6. DS Product Manager
7. Data Vizualizer/CC
2/ As mentioned I’m sure in larger companies even more specific roles exist, but in smaller companies likely you would be a hybrid of such roles. Likely, the typical DS responsibilities will continue to splinter and become more specialized depending on the industry.
3/ I’ll make note of the main distinctions of the most common roles in DS which are—Product Analytics DS, Generalist DS, and MLE
4/ Product Analytics DS must be great communicators with technical and non-technical audiences. They direct the company in such a way to drive KPIs, revenue, etc. This includes reporting, analysis, dashboards, developing metrics, experimentation and simple modeling.
5/ They are typically well versed in SQL, fundamentals of statistics, experimentation, business acumen and communication. Python/R will also be expected but not to the level of a MLE.
6/ Product Analytics DS will optimize existing features, and propose new features of a product. This includes testing variations, determining best fit metrics to optimize for and providing context for MLE to build out models.
7/ Generalist DS are most common and your specific responsibilities will be highly related to the company, department, team that you are on. That being said you should be competent in all aspects of DS. Python/R, stats, experimentation, reporting, communication.
8/MLE roles are very technical and actually lends itself towards software engineering. Some may require experience in Java! Likely, these will utilize deep learning models or custom models for very specific problems. It’s important to be aware of time and space complexity.
Where do you fit in?
TLDR; There are so many different roles within the field of DS! You don’t have to be an expert in every element of DS to be successful. Some roles stress one over the other so find your passion whether it be programming, providing actionable insight or experimentation.
