The Data Science Career Path and Skills Progression

The Data Science Career Path and Skills Progression

Overview

How does the position change as you transition from being a junior data scientist to a senior data scientist?

The data science career path from junior to senior data scientist varies greatly in skill level, responsibilities, daily tasks, and everyone’s favorite topic– total compensation.

For any kind of technical role, including data science, there are two paths: the management path and the individual contributor path.

The individual contributor path includes data scientists who work on core projects, contribute code, run analyses, and build ETL pipelines and machine learning models. The management path encompasses data scientists who manage people, scale data strategy, and work on fitting the pieces of a data organization together.

Both paths originate from the same journey from entry-level to the senior data scientist position, where they then diverge. As career advancement continues, individual contributors can decide to become managers or remain highly specialized data scientists.

For this article, we’re going to focus on the individual contributor data science career path and what it takes to level up as a data scientist.

Data Science Intern & Entry-Level Data Scientist Role

General framework

Let’s start with junior data scientists and data science interns. What do you have to know and what do you actually do? How can your data science career grow?

Honestly, not that much. At this stage, data scientists are super raw and mainly work on developing their core technical skills, like SQL and Python. As such, tasks are generally straightforward and have a clear objective goal.

If you’re a data science intern, you can usually contribute value by building scripts or prototyping projects with data visualizations and models. You won’t have too much time to ship production code or build up an understanding of everything in the business, so the best thing you can do is add value wherever and whenever possible.

Junior data scientists also work within a specific scope. For example, instead of tackling ambiguous analytics problems (e.g., what’s influencing customer purchasing behavior on our e-commerce website?), an entry-level data scientist would more likely be given the task of writing a query to calculate customer churn rates or building a dashboard to look at purchases by marketing channel.

If an entry-level data scientist were to be given a more advanced task of building a model and deploying it into production, there would probably be a daily check-in with senior data scientists to help them get unblocked, do code reviews, and learn how to integrate into the existing system.

Notice the keyword I’ve been using here: junior data scientists are given a task instead of finding one. Understanding what to build separates junior data scientists from more advanced roles. A junior data scientist might brainstorm on strategy and architecture, but the majority of their tasks are focused on producing work for their managers and other stakeholders.

Salaries for data science interns can usually range from minimum wage to $40 per hour at some FAANG companies. Similarly, entry-level data scientists are usually paid around $80k-$100k per year, but it’s pretty common for entry-level data scientists at FAANG companies to make over $150K a year in total compensation, even in their first year. In this case, these companies are hiring for the data science potential more than the actual value these scientists provide.

Mid-Level Data Scientist

Visual of a person in front of computer

After around one to two years of experience as an entry-level data scientist, your data science career path can transition into a mid-level data scientist role. Mid-level data scientists are advanced individual contributors that can take up larger project scopes and more ambiguous business problems.

For example, while a junior data scientist would create the SQL queries for an ETL pipeline, a mid-level data scientist should be able to architect the entire ETL pipeline from scratch and use it in their machine learning model.

A data scientist who has moved past the junior stage won’t need as many check-ins and can usually unblock themselves without asking other data scientists for help. Additionally, from a product perspective, a mid-level data scientist has a higher level of understanding of business problems and how to use data science to solve those problems.

This advancement in skill and experience means more autonomy in terms of project choice and project management. Data scientists will always have more than enough projects to work on. Prioritizing projects is the first step in leveling your career so that someone doesn’t have to assign work to you manually.

Most mid-level data scientists can earn anywhere from $120k-$180k in the Bay Area and other cities, no matter where they end up working. At FAANG companies, this number can shoot up to over $200k a year in total compensation.

Senior Data Scientist Career Path

Four person in a suit standing

Finally, what separates senior data scientists from the rest? While years of experience still matter, there are better features that differentiate a more senior data scientist from a less senior data scientist.

Once someone has worked in data science for five to seven years, their title will likely be denoted as a senior. But someone with 20+ years of experience could still be a worse data scientist than someone with five years of experience. Many companies have now instituted different “levels” to evaluate candidates that are going for individual contributor roles that help determine total compensation as well.

Beyond senior-level positions, many transitions to data science managers. See our Data Science Manager Salaries report to learn more about the role and salary expectations.

The skills of a data scientist can be measured in interviews by testing speed and accuracy on technical problems, but also by evaluating communication skills. Then, on the job, many of these skills are more observable. For example, senior data scientists should be able to:

  • Onboard themselves on business and technical architecture
  • Have high data accuracy and quality
  • Good code quality and completeness
  • Understood project scope and where to prioritize applications of data science
  • Good communication of technical concepts
  • A strong ability to mentor junior data scientists

Scoring well on these different traits determines how senior a data scientist will be. The best data scientists can take a highly ambiguous problem and architect a solution from beginning to end by themselves or in a team environment. The level of efficiency with which they can complete this task determines their value.

For example, let’s say a startup wants to build its first A/B testing system. A good senior data scientist would figure out business requirements and scope–

  • Why do we need an A/B testing system?
  • Do we need it for email, the backend, or only the front-end?
  • How many users does the system need to handle in the future?

Now that the scope is laid out, the senior data scientist would begin architecting a system to build. They would think about how to randomly distribute the users into different buckets, how to create different functions that other data scientists can reuse in their code later on, and what kind of deliverable would be built so that product managers and executives could run experiments and monitor tests.

So how much do they get paid? The short answer is a lot. A senior data scientist, depending on their level, can make anywhere from $150k to a million dollars per year or more. The best senior data scientists understand how they can justify their salaries.

Ultimately, at the end of the day, data science career progression can be like any other role. As human beings, we take on bigger and bigger tasks as we gain experience in the game of life. Your value as a data scientist then corresponds to how much value you can add and the solutions you build relating to data.