OSL Retail Services

Dennis Cheung, Director, Data Engineering and Governance, OSL Retail Services

Challenges and Trends Impacting the Security Space

As data continues to become a core asset to organizations in having a competitive edge, technologies and processes in data security are becoming increasingly sophisticated. That said, the challenges in data security are less dependent on the technologies and ecosystems they are housed in, but rather the people who internally have access to them. As applications continue to be the norm in how businesses and individuals access data, the threat to data security now is the lack of processes and governance in how individuals access and use the data housed within these applications. Given that applications are not created to safeguard data down to its smallest delimitator, the challenge is how then can the proper security roles be created and properly maintained down to each individual who may need to access data for their day-to-day operations? If however, one successfully creates enough security roles for each individual and type of data access, another issue then arises in not only maintaining them; but how can you prevent security roles that typically are grouped: not to grant access to fields that some users may not require.  

Predicaments in the Security Space that Keeps You Up at Night

What keeps me up at night is the fact that applications are not by design, created in a way to safeguard data. As mentioned above, since security groups are usually set by a module within applications – there is no doubt that individuals across various organizations have access to data they should not have. Coupled with the lack of data governance in organizations and individuals themselves do not generally receive the proper data training, the possibility of data being used and or accessed improperly is quite high.

Latest Project and the Technological and Process Elements Used in It

There was a project to move existing data pipelines from on-prem to private cloud for better reliability, scalability, and flexibility for internal and external clients. The technologies that made this project successful given the resource constraints of the team, coupled with evolving knowledge and skillsets meant that we had to utilize technologies that were lower-code but allowed individuals to upgrade their skillset without having deep knowledge in particular tech-stacks. Ultimately it was decided that an ETL tool (KNIME) would be used to help replicate and execute our ETLs. However, given that all new tools require a learning curve; the way we overcame this was through the engagement with external consultants/ service providers who were well-versed in the tool to help translate some of our most complex data pipelines while the internal team would then take the opportunity to learn from these engagements and learn from similar converted workflows to translate and complete the rest in the allocated project timelines set forth.

Exciting Technological Trends for the Future

Traditionally, data warehouses were built to restrict access to data. But with the rise of new technologies like Trino this allows for similar outcomes but without the effort and need to duplicate and or replicate data in other databases or locations to restrict access. In addition, trends in AI are also interesting as technologies and applications themselves begin to incorporate AI to identify potential data breaches and reporting of leaked data.

Brief Overview of the Career Path Leading Up to Your Current Position as Director, Data Engineering and Governance at OSL Retail Services

I have worked in the data domain for over a decade in various roles and organizations. Starting with technology implementation, implementing hospital information systems; then progressing to implementing and creating data standards in the Ontario government. This was the first transition into data management and my data journey. From here, I progressed into various roles in numerous other health-related organizations (public, private, not-for-profit) ranging from data architecture, software & standards development, data governance, data quality, and then ultimately to data engineering. These various evolving roles have allowed me to understand the various domains outlined in the data management association and appreciate how data security can be implemented and play a role across the various data functions.

Adapting to Changes

Not being shy to learn new things, asking ‘dumb’ questions, and knowing when; in a particular role you have stopped learning, has enabled me to continue growing and evolving in my career. In addition, I believe in continued learning and professional certifications in order to keep the designation – this encourages you to stay on top of industry topics and best practices. Lastly, being a professor has also helped me stay on top of the industry as each year requires a revisit in content for current use cases and trends. In addition, students are also very insightful as they will bring up topics/ trends that are of interest to them sparking my inquiry into those domains and technologies.