Different DS career paths

  • 13 April 2022
  • 0 replies
  • 24 views

Userlevel 2
Badge

Data science is a broad, ill defined, and ever growing field. Out of that mess, though, more defined career paths are slowly arising. These are the 7 most common that I have picked up on - in no way a complete list, or even a good one, but I think it’s good to know what your options are. I’ve tried to order them from more code focussed to more people focussed, but all roles will require skills in both.

Data engineering involves getting the data into the right place at the right time, and in the right format. It’s crucial for any sustainable solutions. Data engineering is the more structured side of this, most of them come from a compsi background and you need qualifications.

Machine Learning engineers are often your more advanced ‘data scientists’, and they work on deploying algorithms that scale with volume and deliver outputs quickly. A lot of ML engineers are also involved with research, although it isn’t a requirement. Most of the DS that work at the likes of Deepmind, OpenAi, or Vicarious, where you’re doing long term research on cutting edge, either currently are, or have at some point been called an ML Engineer. 

Research and Development: typically what you’ll be doing here is working on tooling. So generalising solutions and building tools for your team to use in your company, or open-sourcing them. 

Operations based data scientists: This is typically end to end project work from designing the problem, building the models, implementing and testing the solution and productionising what you’ve built. It can often strike a good balance between working in teams/with customers, and actually building things. 

Insight data scientist. I’d say the difference here is that you’re focussed more on quickly finding and presenting insights from a business’s data than developing complex models. Often the most value you can provide to a business is from good analytics of their data, especially when a business doesn’t have good data maturity. A lot of customer segmentation, for example, can be done without touching an ML technique. 

AI consultants: their role is exploratory opportunity analysis. So they use business domain knowledge, as well as their technical experience, to discover areas where businesses can improve processes using data science, ML, or AI techniques. 

Then what happens with a lot of people who come into this field, is once they have experience, they become team leads. You’re not really touching any code anymore, you’re more into people management, or project management, which you can do now with your practical experience.

 


0 replies

Be the first to reply!

Reply