What does being a leader entail for an industry data scientist? For some practitioners, leadership means managing people and products; for others, the focus might be on developing workflows and practices that help companies succeed. It can even be both! Regardless of the specific definition of your role, your work likely requires navigating fast-changing demands alongside peers and stakeholders across multiple disciplines. It’s not an easy skill set to acquire, but fortunately it’s one we can cultivate.
This week, we’ve selected several recent posts about leadership that combine high-level principles and pragmatic insights. Regardless of the type of Data Science leader you are, they will inspire you to reflect, reconsider your approach, and experiment with new tools and ideas. Let’s dive in.
- It’s crucial to acknowledge that leadership is hard. Popular culture will have us believe that leaders are synonymous with certitude and decisiveness. Chris Walsh‘s post is a much-needed antidote to that perception. It invites us to work through the discomfort that comes with a decision-making position, and addresses the specific challenges data-science managers face.
- Few things matter more than setting a strong foundation. For Salma Bakouk, building a data team from scratch was a process of trial and error, and external guidance was scarce. She shares the lessons she learned along the way—especially around keeping a strong alignment with the company’s objectives and stage of data maturity—and provides a roadmap for people who are tackling a similar task.
- The importance of complementary skills. Addressing a similar problem from a different direction, Marie Lefevre offers concrete advice for practitioners responsible for developing a company’s data stack (and hiring the people who will implement it). For Marie, a well-run data operation balances the skill sets of data engineers, data analysts, and data managers.
- The qualities that help data leaders stand out. As Google’s Chief Decision Scientist, Cassie Kozyrkov has had the opportunity to think deeply about what makes some data analysts great while others struggle to excel at their job. In her latest article, she distills her thoughts into 10 distinct elements you can work on to grow and move forward in your career.
- Preparing yourself (and your team) for change. In a fast-evolving field, being an effective leader means you can tell when the terrain is shifting—and you know how your team should adapt. Barr Moses believes we’re moving away from the dominance of big data, and offers actionable ideas for data pros who want to ensure their stack and their workflows are ready for the future.
- Growth means taking feedback seriously. Jason Chong recently marked his first anniversary as a graduate data scientist, and wrote about the experience of going through his inaugural performance review. His takeaways apply to people up and down the org chart, though: he stresses the key skill of listening to feedback, including (if not especially) the bits that make us aware of our shortcomings.
If our collection of articles on Leadership has left you in a reflective mood, fantastic: take your time to digest. If you’re still in the mood for some excellent reads on other topics, well, here you go:
- No recent AI development has taken social media by storm quite like DALL·E 2. If you wanted to learn more about the promise and limitations of the image-generation model, Alberto Romero‘s deep dive is the one-stop resource you need.
- "The changes we are experiencing today are unquestionably altering the map we use to organize our knowledge of the world." Prerna Singh explored how the evolution of data science and the availability of massive datasets affected the field of global studies.
- If you’re in the early stages of your NLP learning journey, you’ll want to check out this thorough, hands-on tutorial on the spaCy library, presented by Bex T.
- Add depth to your understanding of ML infrastructure with Sivan Biham‘s introduction of an end-to-end pipeline solution for training machine learning models, from containerizing to tracking.
To all our readers: we appreciate your support, your passion for data science and good writing, and your decision to be part of our community. Special thanks go to those of you who’ve taken the leap and become Medium members, which helps us and our authors share more great with you.
Until the next Variable,
TDS Editors
