Becoming a Level 3.0 Data Scientist

By Jan Zawadzki, Project Lead Data & AI at Carmeq GmbH

Companies are hiring Data Scientists on three levels: Junior, Senior, or Principal. Whether you’re just getting started with Data Science or looking to switch careers, you will inevitably find yourself on one of these levels.

This post aims to shed light on what’s expected and what’s outside of the scope of each Data Science Career Level. While companies might have different job titles, this post provides a general baseline. Furthermore, the post concludes with hands-on tips on how to prepare your career for the switch into AI or the well-deserved promotion.

Let’s level up.

 

The Data Science Skills Matrix

 
A Data Scientist is expected to have knowledge in three areas: Statistics, Engineering, and Business. However, you’re not expected to master all three areas right from the get-go. Which skills should you focus on when seeking an entry-level position? Which skills become more important as you progress through the career ladder?

The graphic below shows the market expectations for each Data Science Level from 1.0 to 3.0. The results are based on my personal experience in the field and conversations with experts and influencers from Stanford, eBay, Axel Springer, and Xing. To avoid confusion, we will refer to the positions as Level 1.0 to 3.0.

Let’s analyze the expectations of these positions in detail below.

 

The Junior Data Scientist — Enter Level 1.0

 
The prototypical Junior Data Scientist is a young graduate. Popular fields of study include Computer Science, Mathematics, or Engineering. A Junior Data Scientist has 0–2 years of work experience and is familiar with creating prototypes with structured data sets in Python or R. She has participated in kaggle competitions and has a GitHub profile.

Junior Data Scientists can provide tremendous value to companies. They are fresh off taking online courses and can provide immediate help. They are usually self-taught since few universities offer Data Science degrees and thus show tremendous commitment and curiosity. They are enthusiastic about the field they have chosen and are eager to learn more. The Junior Data Scientist is good at prototyping solutions, but still lacks proficiency in engineering and business mindset.

Junior Data Scientists should have a strong passion for Machine Learning. You can demonstrate your passion by contributing to open-source projects or participating in kaggle challenges. — Dat Tran, Axel Springer AI Lead

 
What they do

If a company is hiring Junior Data Scientists, usually a Data Science team is already in place. The company is then looking for help to make life easier for more experienced colleagues. This involves rapidly testing new ideas, debugging, and refactoring existing models. You will discuss ideas as sparring partners with the team. You pitch new ideas on how to do things better. You take responsibility for your code, continuously striving to improve code quality and impact. You’re a great team player, constantly looking to support your teammates on their mission of building great data products.

 
What they don't do

Junior Data Scientists don’t have experience in engineering complex product solutions. Ergo, she works in a team to put Data Science models into production. Since the Junior Data Scientist just joined the company, she is not immersed in the business of the company. Hence, she is not expected to come up with new products to impact the Fundamental Business Equation. However, what is always expected is the desire to learn and improve your skills.

I’m interested in a Junior Data Scientists ability to complete a non-trivial project. By complete, I mean that the project was completed from start to end by the person — or within a group — and culminated in a fully flushed-out product. I find it to correlate with the Data Scientist’s ability to lead projects on the job. — Kian Katanforoosh, Founding member deeplearning.ai, Stanford CS Lecturer

If you excel at being a Junior Data Scientist, you have a strong background in understanding Data Science models. You show insatiable curiosity to learn about engineering and the business of your company to improve your skill set.

What does the next level look like? Let’s find out about Senior Data Scientists next.

 

The Senior Data Scientist — Reaching Level 2.0

 
The Senior Data Scientist has already worked as a Junior Data Scientist, Software Engineer, or completed a Ph.D. He has 3–5 years of relevant experience in the field, writes reusable code, and builds resilient data pipelines in cloud environments.

Senior Data Scientists should be able to frame Data Science problems. Good candidates have great insights from past Data Science experiences. I also dig deeper into their ability to write production code. — Kian Katanforoosh

Companies prefer to hire Senior Data Scientists because they provide tremendous value at a reasonable salary. They are more experienced than Junior Data Scientists, thus omitting costly greenhorn mistakes. They are also not as expensive as Principal Data Scientists, while still being expected to deliver Data Science models in production. It’s a very fun level to play, having surpassed Level 1.0 and yet having room to grow to Level 3.0.

 
What they do

The Senior Data Scientist masters the art of putting mathematical models into production. While Principal Data Scientists or Business Managers assign tasks, the Senior Data Scientist takes pride in building well-architected products. He avoids logical flaws in the model, doubts systems that perform too well, and takes pride in preparing data correctly. The Senior Data Scientist mentors Junior Data Scientists and answers business questions to management.

 
What they don't do

The Senior Data Scientist is not expected to lead entire teams. It is not the responsibility of the Senior Data Scientist to come up with ideas for new products since they are generated by more experienced colleagues and managers. While the Senior Data Scientist knows the details of the products they have built, they are not expected to know the overall architecture of all data-driven products. The Level 2.0 Data Scientist is skilled in statistics and better in engineering than a Level 1.0 Data Scientist but strays away from the non-fun business part from Level 3.0.

A Senior Data Scientist has to be able to bring their code in production (with some support of the data engineers). Seniors should be able to complete the projects they are given independently. — Sébastien Foucaud, VP of Data Science at Xing

The Senior Data Scientist is measured by the impact their models generate. He has a good intuition about the inner workings of statistical models and how to implement them. He is in the process of understanding the business of the company better but isn’t expected to provide solutions to business problems just yet.

Now that we’ve investigated Level 2.0, let’s see what the final Level 3.0 looks like.

 

The Principal Data Scientist — Endgame Level 3.0

 
The Principal Data Scientist is the most experienced member of a Data Science team. She has 5+ years of experience and is well-versed in various types of Data Science models. She knows the best practices when putting models to work. She knows how to write code computationally efficient and is lurking around to find high-impact business projects.

In addition to her impeccable engineering skills and deep understanding of the scientific models used, she also firmly understands the business that her company works in. She has a track record of impacting the business baseline with Data Science.

Principal Data Scientists need to have a very good understanding of the business problem they are solving before writing one line of code. Meaning, they need to have the ability to validate ideas prior to implementation. This approach increases the Data Science project success. — Adnan Boz, AI Product Leader at eBay

 
What they do

The Principal Data Scientist is responsible for creating high-impact Data Science projects. In close coordination with stakeholders, she is responsible for leading a potentially cross-functional team in providing the best solution to a given problem. Hence, her leadership skills have developed since Level 1.0 and 2.0. The Principal Data Scientist acts as a technical consultant to Product Managers from different departments. With her vast experience and skills in the major Data Science categories, she becomes a highly valued asset to any project.

 
What they don't do

While shaping the discussion about desired skills, it is not the responsibility of the Principal Data Scientist to recruit new team members. Although she understands the business of her company and suggests impactful new products, the Product Managers are still responsible for market adoption. She also leads teams, but career progression decisions are still taken by the team lead.

The Principal Data Scientist should steer projects independently from the Head of Data Science. This person is expected to obtain first leadership skills and therefore it is important that this person communicates clearly, is empathetic, and has a good eye for people. — Dat Tran

The Principal Data Scientist has seen why products fail and thus she drives new projects successfully. She is a valued contributor to product discussions and enjoys educating the company about Data Science. With her experience in delivering impactful Data Science solutions, she is the most valuable asset in the Data Science department.

Now that you’ve seen the different expectations of a Data Scientist from Level 1.0 to 2.0 until 3.0, let’s find out how you can use this knowledge to advance your career.

 

Leveling up your Data Science Career

 
It doesn’t matter if you’re looking to enter the Data Science Career Game on Level 1.0 or you’re looking to progress into a higher Level. Take the following steps to make your next career move.

 
1. Evaluate your skills

As a first step, compare your skills to the Data Science Skills Matrix. How solid are your statistics skills? How good are your engineering skills? How business-savvy are you?

 
2. Plan your promotion

Many companies have annual promotion cycles to advance their employees. Julie Zhuo, VP of Product Design at Facebook, recommends as the first step to get promoted to make your aspirations to level up known. State at the beginning of the progression cycle that you’d like to advance your career to your manager. Ask your manager how she ranks your current skill set and what is expected to reach the next Data Science Career Level.

 
3. Improve your skills

Once you’ve analyzed your skills and announced your desire for a promotion, it’s time to level up your skills.

Do you want to break into AI? Then get rock solid about statistical models and learn how to solve problems with structured datasets. Do you look to enter Level 3.0? Make sure you have your math, engineering, and business skills on lock. Armed with these points, you’re well-prepared to negotiate the promotion with your manager.

 

Key Takeaways

 
Navigating the Data Science Career Levels is fun. Remember the following Key Takeaways:

• Junior Data Scientists have good statistical skills
• Senior Data Scientists excel at putting models into production
• Principal Data Scientists know how to create business value
• To level up, evaluate your skills, announce your desire to progress, and work on outstanding skills

Nuances exist within the different Data Science Career Levels. For instance, Séb Foucaud is rather looking for strong engineering than math skills in Junior Data Scientists. Some Senior Data Scientists might discover their passion for building scalable data pipelines and transition to a Data Engineering role. Some Principal Data Scientists prefer to develop technical expertise while others rejoice in focusing on business skills. Whatever career path you take, developing your skills around the three main areas of Data Science expertise will get you far.

This post continues the ongoing series of educating Data Scientists to become business-savvy. The series aims at helping you polish your overall Data Science skill set. If you enjoy the format, please follow me on LinkedIn or Medium to stay up to date with new articles.

 
Bio: Jan Zawadzki has 5 years of experience at a global consultancy and as a Data Scientist. Jan currently works in the realm of self-driving cars as a Project Lead Data Science for Carmeq GmbH, the innovation vehicle of Volkswagen AG. Jan is passionate about advancing the automotive industry through machine learning and sharing his knowledge in the fields of Business and Data Science. He is a monthly Contributor to the “Towards Data Science” Publication on Medium. He is also a Deep Learning.ai Ambassador, supporting the team around Deep Learning Luminary Andrew Ng.

Original. Reposted with permission.

Related:

• The Third Wave Data Scientist
• 5 Things to Review Before Accepting That Data Scientist Job Offer
• Interview Questions for Data Science – Three Case Interview Examples