Data Scientist with Seniority is not the same as Senior Data Scientist

Skills and Experience don’t automatically translate into Seniority. Which has more to do with behavior and attitude.

We often tend to define Seniority as a measure of years of experience, things you are able to do (hard skills), or social competencies (soft skills). When it comes to Data Science, demonstrating Seniority is a little bit different than that. It has to do with attitude, accountability, and how well you are able to sync with surrounding environments.

Photo by Austin Chan on Unsplash

This is my personal guide on what behaviors you should look for when deciding which of your Data Scientists would make a good leader.

1. Integrity is the foundation of everything a Senior Data Scientist does. A Data Scientist with Seniority doesn’t “go get some data” cherry-picking a sample (sampling bias) or omitting some key facts (omission bias), to prove the boss’s gut feeling. Doesn’t visualize data in ways that are misleading, just to prove a point. Understands and avoids the major bias and paradoxes in data. Doesn’t use data in ways that are harmful to other people and follows the law when it comes to data privacy. In other words, is conscious that Reputation is their best career asset.
2. Has got ego under control. When taking a decision on how to approach a problem, only favors the solution that best aligns with the company’s technology and objectives. Doesn’t come up with a Trasformers-based architecture for a time series problem just because they want to learn Trasformers. Chooses ARIMA if it gives good results and can be deployed to production in no time, without even giving it a second thought. Doesn’t make a fuss about using Java, if everybody in the team is using Python. Doesn’t say “I’ve got a Ph.D.” every three words.
3. On the flip side, has come to terms with Impostor Syndrome. Doesn’t live in eternal self-pity for the many things they don’t know. Knows how to make the most of their strengths and acknowledges weaknesses. It’s ok not to have a Master's in Machine Learning at Stanford, many very successful Data Scientists have learned by doing.
4. Can work independently. Doesn’t need the boss to hold their hand or sign off on every decision they make or approve any line of code. Doesn't need a 25-page functional analysis that describes in every detail what to do, but can get started from an idea sketch on a clipboard. Delivers results with little supervision, expanding and improving on the initial idea along the way. When assigned a project, owns it and feels responsible for it. Doesn’t need to be asked for updates, but will knock at their boss's door to provide some updates.
5. Takes pride in their ability to successfully communicate and explain complex ideas to others. This is kind of a corollary of point 2. A Data Scientist with Seniority and Ego well under control doesn’t put a tonne of math and jargon in PowerPoint slides just to intimidate non-technical team members. In fact, has the intelligence to tweak the communication style to the audience. For example, they may use technical language when presenting to fellow Data Scientists, but simplify their language for non-technical colleagues. When presenting to Senior Executives, they may focus on high-level business metrics and outcomes. Doesn’t get frustrated if a finance manager that has seen bar charts and pie charts all his life doesn’t understand boxplots and quantile plots in a 10 minutes presentation. Most of all doesn’t conclude that the Finance Manager is someone stupid and not worth talking to.
6. Creates a pleasant working environment, and is someone nice to work with. Encourages other team members to work together when it’s needed, openly shares ideas and knowledge (mentoring, training), and collaborates on projects. Doesn’t discredit other people’s idea just because it’s different than their own idea. When asked for advice on reviewing some Python code doesn’t rewrite it in Go because “it’s faster like that”. Recognizes and appreciates other team members’ hard work and achievement, is positive and supportive of others.
7. Understands how their company makes revenues. This serves as a guiding principle when deciding between competing projects or solutions, ensuring that personal preferences or ego do not cloud judgment. Which project/solution best fit the company's goal and objectives? It is not uncommon to encounter Data Scientists who have worked for a company for several years and yet have little understanding of how the company makes the money to pay their salary. Very often they are those who underperform or quit after a short-term project.
8. Last but not least, A Data Scientist with Seniority is a lifelong student who understands that the calculus exams they did 10 years ago are a great background but may be not enough to be on top of the game. On the other side, knows when it’s time to pause the “from zero to hero" course to go study some calculus. Data Science and Machine Learning evolve rather quickly. For example, the Attention Mechanism, which made innovations like ChatGPT possible, is less than a decade old. Staying up-to-date with the latest advancements and technologies requires a constant and dedicated effort and is essential to be able to provide the most valuable insights and solutions. No slack allowed.

Photo by Becca Tapert on Unsplash