From DBAs to Analytics Engineers: The Future of Data Roles

youssef mjoune
Sep 28
3 min read

Introduction: Why Data Roles Matter

Over the last thirty years, the story of data has been told through its roles. From Database Administrators (DBAs) to Analytics Engineers, each profession reflects how organizations prioritize stability, agility, innovation, or governance.

Roles are more than job titles. They are indicators of pain points, opportunities, and cultural shifts. Understanding t

heir history helps us anticipate the future of data jobs in an AI-driven world.

The 1990s: Database Administrators as Gatekeepers

In the 1990s, the enterprise world migrated from mainframes to relational databases. Oracle, IBM DB2, Sybase, and SQL Server dominated.

The role: The DBA was the guardian of corporate data.
The focus: Stability, integrity, and security of databases.
The trade-off: Data was safe but inaccessible. Business users had to wait weeks for reports.

This was the first form of data governance — one built on control rather than enablement.

The 2000s: BI Developers and Data Warehouse Architects

The 2000s were the era of Business Intelligence. Companies invested millions in data warehouses to chase a “single version of the truth.”

BI Developers built dashboards in tools like Cognos and Business Objects.
Data Warehouse Architects modeled star schemas and designed ETL processes in platforms like Informatica or DataStage.

Impact: Executives gained visibility through dashboards. But schemas were rigid, and agility suffered. Still, this era gave us dimensional modeling discipline and the first direct collaboration between technical teams and business stakeholders.

The 2010s: Data Scientists and Big Data Engineers

The explosion of unstructured and semi-structured data defined the 2010s. Logs, clicks, social media, images, and sensors created a new reality: Big Data.

Data Scientists emerged as hybrid experts in statistics, coding, and domain knowledge, building predictive models in Python, R, and Jupyter.
Big Data Engineers set up Hadoop clusters, Spark jobs, and Kafka pipelines to make massive datasets usable.

Lesson learned: Without production discipline, models stayed in notebooks. Organizations realized that data science without engineering is fragile.

The 2020s: Analytics Engineers and Data Product Managers

The 2020s corrected the excesses of the big data era. Cloud-native tools like Snowflake, BigQuery, Synapse, and Databricks brought elasticity and simplicity. At the same time, dbt brought software engineering practices to analytics.

Analytics Engineers deliver tested, documented, versioned datasets.
Data Product Managers treat data like a product, focusing on usability, adoption, and lifecycle management.

This decade marks the rise of data as a product, measured not just by ingestion but by adoption and value.

The Future: AI Data Stewards, Prompt Engineers, and Hybrid Roles

The acceleration of AI shifts the bottleneck from access to trust and governance.

AI Data Stewards will ensure datasets are ethical, compliant, and explainable.
Prompt Engineers will design effective interactions with generative AI systems.
Data Product Owners will manage adoption and value creation across domains.

Three Scenarios for 2035

AI Everywhere, Roles Converge → governance dominates, technical depth matters less.
Data Mesh Maturity → roles specialize by domain with federated ownership.
Human-AI Symbiosis → hybrid professions like AI Risk Officers and AI Analytics Architects emerge.

The most likely path? A world where data roles blend human judgment with machine intelligence, creating new professions we can barely name today.

Conclusion: Designing the Future of Data Work

From DBAs in the 1990s to Analytics Engineers in the 2020s, each decade has corrected the failures of the one before. Stability, visibility, prediction, and now balance — every role carries the culture of its time.

The next frontier is not only technical but human. AI governance, ethical data stewardship, and human-AI collaboration will define the coming decades.

If we want to thrive in this future, we must design not only architectures and platforms, but also the roles and responsibilities that shape how people interact with data.