The role of database administrators (DBAs) is changing rapidly. Whereas data used to be neatly stored in database systems, it can now be found everywhere: in Parquet files, Excel spreadsheets, and countless cloud environments. Meanwhile, companies are struggling with security, automation, and an acute shortage of experienced specialists. We spoke with Redgate Ambassadors Kellyn Gorman and Tonie Huizer about the reality of database management and how the DBAs are adapting to a world where data no longer stays within clear boundaries.
The democratization of data has brought many benefits to organizations, but it has also made DBAs’ work more complex than ever. Information moves freely between applications, storage layers, and users, quickly blurring the view of data and access rights. Kellyn and Tonie see that most companies are struggling with this development. No one knows precisely who has access to which data or who enters what into which system.
The situation is becoming even more challenging with the rise of AI and advanced analytics. Many organizations want to innovate quickly with data and models but are skipping crucial governance and policy steps. The result is often a chaotic landscape in which data spreads before the right controls are in place.
Redgate Ambassadors are experienced data experts who work closely with Redgate to share knowledge and support the development of new solutions. They are part of the global Redgate community and contribute to improving tools, creating learning content, and strengthening the data ecosystem.
Governance often comes too late
Companies typically prioritize innovation over governance. New environments in Databricks or Snowflake are set up to work with data quickly, while policies and responsibilities are worked out later. As a result, it is often unclear who is liable in the event of incidents or data breaches.
Kellyn and Tonie cite last year’s Snowflake breach as an example. Many users did not even know they had been affected. We must take joint responsibility, say the two Redgate Ambassadors. Database security is not an individual task, but a shared responsibility that requires cooperation between all parties involved.
One of the core principles of modern database security is observability: the ability to see exactly who has access to which systems, tables, or datasets. That sounds obvious, but in practice it proves difficult. For example, someone could access sensitive data via an underlying folder without it being visible in the access logs.
Redgate responds to this with solutions such as Redgate Monitor, which provides insight into user rights, system access, and health checks. This allows DBAs to quickly see who has access to what and which privileges may have been set too broadly. This insight forms the basis for a secure data infrastructure, especially in hybrid environments where multiple clouds and databases converge.
Redgate Flyway bridges developers and DBAs
Automation and deployment are another important focus area. Flyway, part of Redgate for several years now, plays a key role here. The 2019 acquisition was a step for Redgate to expand beyond Microsoft SQL Server and Oracle. The Redgate Flyway tool now supports more than 20 database systems and enables versioning of database changes. This means that changes are checked and rolled out step by step from development to testing and production.
This approach prevents unseen changes from being implemented that could cause security issues. For example, when someone adds a new column to a table, this change is first reviewed by colleagues. For Oracle DBAs, this means they can immediately see when someone tries to assign overly broad privileges, such as ‘grant alter session’. Early checks prevent vulnerabilities from reaching production.
Entity Framework requires additional checks
A notable point of discussion within database management is the role of Entity Framework, a Microsoft tool that enables developers to generate tables and objects without extensive database knowledge. For DBAs, this is often a source of frustration because the output does not always meet performance and security requirements.
For this reason, Redgate Flyway is increasingly positioning itself as a bridge between developers and database professionals. Every change made via Entity Framework appears as a pull request in Redgate Flyway, allowing DBAs to review it before it goes into production. “Why is this table being modified? Why is a new column being added?” are questions that are then asked.
By automating this process, control is maintained without compromising software development speed. This creates a balance between agility and security, which is exactly what modern organizations strive for.
Automation as a lever in the face of a shortage of DBAs
Automation is not just a means of preventing errors. It is also a necessity in a tight labor market. “Personally, I don’t want to do anything manually more than once,” says Tonie. By automating repetitive tasks, more time is left for strategic work, such as optimization, capacity planning, and security audits.
Yet many organizations do not yet see automation as a top priority, even though it is one of the most effective ways to address the shortage of DBAs. By standardizing processes and eliminating manual steps, teams can work more efficiently and securely.
The myth of the disappearing DBA
Articles regularly appear suggesting that the role of the DBA is disappearing due to AI and automation. According to Kellyn and Tonie, this is a misconception. The function is certainly not disappearing, only the title is changing, they argue. For more than ten years, people who perform DBA tasks have often been called data engineers, site reliability engineers, or cloud architects. But they are still essentially doing database management, Kellyn and Tonie note.
It remains human work. Developers think in terms of functionality and requirements. DBAs, in turn, look ahead to optimization, stability, and security. AI or scripts cannot simply replace that difference in mindset. Professionals who understand how data actually flows through systems remain indispensable to the reliability of any organization.
Another interesting point that came up during our conversation is that many organizations often don’t even know they are running multiple databases. A company that thinks it is running entirely on Microsoft may in fact also be using Postgres via a cloud service. This trend is reinforced by cost awareness and the flexibility of open source solutions.
Tip: Redgate database management strengthens the bridge between people and technology
AI: promise and challenge
No conversation about database management in 2025 is complete without AI. Both Kellyn and Tonie use AI tools daily to speed up their work. For example, Tonie uses Codex to quickly get answers to questions about documentation in his GitHub repository.
The possibilities are endless: AI can generate test data, perform code reviews, and check data quality. The latest version of Redgate Test Data Manager can even automatically create realistic test data. Instead of a hundred instances of “John Doe,” you get a hundred unique names or company profiles, depending on your prompt, to illustrate an example.
Nevertheless, Kellyn and Tonie emphasize that caution is advised. Kellyn experimented with a local LLM via Ollama, which assesses DDL changes in Redgate Flyway. “I deliberately work offline so that data remains secure within the organization,” she says. Her credo is clear: “Only use AI if it really adds value, not because you can.”
Data as the foundation for everything
Amidst all the changes, one truth remains: transactional databases are the foundation of every digital organization. “Without transactions, there is no analytics,” Kellyn summarizes. Many companies focus on trends and dashboards, but forget that the reliability of those insights depends on the quality of the underlying data.
The role of the DBA may be changing, but its importance is only increasing. Whether they are called data engineers, SREs, or cloud architects, these professionals lay the foundation for AI, analytics, and innovation to flourish. Database management in 2025 is no longer just about tables and queries, but about insight, responsibility, and the ability to bring order now that data is everywhere.