Financial Services has taken 20 years to recognise that it is a software business. It does not have another 20 years to recognise that it is enabled by data. Unlike software natives or data natives, traditional businesses are learning and retrofitting practices in tandem with their service or product offering.

The rise of the Chief Data (and Analytics) Officer in mid-2010, was due to the need to safeguard data and evidence control over its usage. This was rapidly followed by traditional businesses wanting to extract value from data to stay competitive and relevant. As we have become data and digital citizens, getting data right is a social responsibility.

For professionals in this field, getting decisions based on data right is a moral obligation.

Since Dodd-Frank and S-OX regulations in the early 2000s, developers have been kept very separate from live operations. The disciplines of DevOps made it easier to protect production environments through engineering rigour and advanced practices and tools. The emergence of analytics professionals in mid 2010 has been mistakenly bundled with software development.

For data scientists, data engineers and data management professionals, real data is their raw material and their product. They happen to exhibit coding skills to build those data products. New interactions with IT teams are emerging (with new team topologies) and with those, new practices, tools and technologies that enable DataOps.

This session is about accelerating understanding and proficiency in the field of data and analytics. In highly regulated geographies and industries, every single outcome of a decision (increasingly automated) must be within the boundaries of the law and - in retrospect - meeting consumers' needs. For that to happen, we must bridge the chasm between software engineers and analytics professionals.

What is DataOps? It is the practice of orchestrating human and automated activities as the data flows through a software-enabled production line, in order to guarantee the integrity of data and decisions based on data. The stations in this production line may be discrete applications or a complex system of people and software. Conceptually, these flows are data pipelines in a data factory. DataOps wraps new disciplines around DevOps, such as the interactions with customers (iterative model design) to operations (statistical process control for drift/bias detection).