The Evolving Role of the Modern Data Practitioner
In the ever-expanding world of data science, the landscape has changed dramatically over the past two decades. Once defined by statistical models and SQL queries, today’s data practitioners must navigate a dynamic ecosystem that includes cloud computing, software engineering best practices, and the rise of generative AI. In a recent episode of ODSC’s Ai X Podcast, Marck Vaisman, a cloud architect and data science practitioner at Microsoft, shared his insights on the evolution of data science and what it means to be a data practitioner in 2025 and beyond.
You can listen to the full podcast on Spotify, Apple, and SoundCloud.
From the Early Days of Data Science to Today’s Complex Ecosystem
Marck’s journey into data science began nearly 20 years ago when the field was still in its infancy. His experience started with basic SQL queries and Excel spreadsheets, but as datasets grew larger, so did the need for more powerful tools. In the early 2010s, the rise of Hadoop and cloud computing transformed the industry, introducing data practitioners to new challenges in scalability and infrastructure.
During this period, Marck became actively involved in the data science community, co-founding Data Community DC and leading meetups to discuss best practices and emerging technologies. Teaching himself R through meetup discussions, he saw firsthand how peer learning played a crucial role in shaping data careers. He has since expanded his expertise into cloud computing and generative AI, working directly with Microsoft customers to integrate AI-driven solutions into their workflows.
Beyond the “Data Scientist” Label: A Broader View of a Data Practitioner
One of Marck’s key takeaways is that the term “data scientist” has become overloaded. Originally framed as a fusion of programming, statistics, and domain knowledge, the role has since evolved to include a diverse set of specializations. He prefers the term “data practitioner” to better capture the broad skill set required today.
He identifies several key specializations within modern data science:
- Data Science & Analysis: Traditional statistical modeling and machine learning applications.
- Data Engineering: The infrastructure and pipeline work that supports AI and data science.
- Data Management & Governance: Ensuring data quality, compliance, and security.
- Research & Project Management: Applying scientific methods and overseeing large-scale data initiatives.
- Generative AI & LLM Applications: A new category focused on leveraging pre-built AI models for automation and augmentation.
For those entering the field, he emphasizes that foundational data skills — such as data wrangling, visualization, and computational thinking — are essential regardless of specialization. Moreover, the ability to adapt to new tools and technologies is more critical than ever, as the landscape continues to shift with the advent of LLMs and AI automation.
The Rise of Generative AI and Its Impact on Data Science
Marck acknowledges the massive impact of generative AI and LLMs on the data science industry. While these technologies are changing how organizations approach problem-solving, he cautions that they don’t replace traditional machine learning or data science workflows. Instead, they serve as powerful tools that can augment human capabilities.
For aspiring data practitioners, understanding generative AI concepts — such as prompt engineering, model fine-tuning, and API integration — is becoming a valuable skill. However, he warns against assuming that AI alone can solve all data problems, emphasizing the need for human oversight, validation, and ethical considerations in AI deployment.
Soft Skills: The Non-Automatable Edge
Beyond technical expertise, Marck stresses the importance of soft skills — what Google’s Cassie Kozyrkov calls “non-automatable skills.” These include communication, teamwork, problem-solving, and the ability to translate technical findings into business impact. While AI can automate many tasks, it cannot replace the human ability to think critically, navigate ambiguity, and collaborate across disciplines.
He advises newcomers to focus on adaptability and continuous learning. Whether working in a large organization with specialized roles or a startup where one must wear multiple hats, the ability to learn new tools and approaches is invaluable. “The tools will change, but the core concepts remain,” he notes.
Breaking into the Industry: Advice for Career Changers
Many professionals are looking to transition into data science from backgrounds in software engineering, business analytics, or IT. For career changers, Marck suggests focusing on three key areas:
- Developing a strong foundation in data skills: Learn SQL, Python, or R, and understand basic statistical and machine learning concepts.
- Building a portfolio of real-world projects: Employers want to see applied skills, not just coursework. Kaggle competitions, open-source contributions, and personal projects can help demonstrate capabilities.
- Networking and community involvement: Engaging with local meetups, conferences, and online forums can provide invaluable mentorship and job opportunities.
He also highlights the importance of setting realistic expectations. While entry-level positions may require familiarity with generative AI, companies should recognize that mastery comes with time and experience. Hiring managers, in turn, should look beyond rigid technical checklists and assess a candidate’s ability to learn, adapt, and contribute meaningfully to a team.
Conclusion: The Future of Data Science Is Flexible
As data science continues to evolve, so too must its practitioners. Whether specializing in generative AI, data engineering, or traditional analytics, the key to success lies in maintaining a growth mindset. Marck’s insights remind us that while the tools and technologies will change, the fundamentals of working with data — problem-solving, curiosity, and adaptability — will always be in demand.
For those looking to stay ahead as a data practitioner, embracing lifelong learning and community engagement will be critical. Whether you’re a seasoned professional or just starting, the path to success in data science is not linear — but it is certainly full of exciting opportunities.
