AI and Data: 5 Trends Affecting the Role of Data Scientist
In today’s data-driven world, a lot of organizations are investing massively into AI strategy and execution, putting the role of a data scientist in heavy demand.
With the growing discussions aroundbig data vs data science, the ascendancy in the need for data scientists has been on the climb for a while now. The Harvard Business Review in 2012, infamously declared the role as “the sexiest job of the 21st century.”
Also, the U.S. Bureau of Labour Statistics (BLS) more recently, published six-figure salaries and thousands of new positions for the coming years. However, with this increasing demand, here’s a look at top trends that are reshaping the data science role.
1. More emphasis on problem-solving
Having data skills is just the minimum requirement of what a data scientist needs to be successful. As businesses begin to grow, data scientists are now required to bring to the table more domain expertise, communication skills and innovative mindset to the whole AI conversation.
“Data scientists generally have gone through a lot of training and practice in AI and ML algorithms, software and infrastructure. But ultimately data scientists’ goal is to solve problems to improve their company’s products and business,” saidYinyin Liu, principal engineer and head of data science at Intel’s AI Products Group in Forbes.
“Before they apply data science, data scientists often need to start with creatively defining the data science problem to achieve that business goal.”
2. The need for more AI most of the time
In the past, data science and AI/machine learning weren’t roped into one. However, that is rapidly changing as more organizations are beginning to adopt the use of AI solutions and need qualified professionals to ensure that they churn out perfect algorithms and AI strategies.
Therefore, as a data scientist, there is the need to have a deep understanding of artificial intelligence—and today, with the option of running an online MBA program, there is really a plethora of ways to acquire this skill.
“Data scientists are now equipped with new methods from the latest AI research, and more powerful and efficient models become available every day,” said Liu. “Data scientists need to stay connected with AI research and customize and adapt these models to specific use cases. In our own data science work, the challenges we face in applying AI to use cases has inspired further research and innovation as well.
“In addition, in building AI solutions it’s critical that data scientists understand the necessary infrastructure and corresponding metrics. That includes whether your company has the necessary data pipeline in place, or whether you have the right computing back-end and resource allocation mechanism to ensure the required bandwidth, latency and throughput for these AI workloads.”
3. The expectations are higher
Like the saying, “with great power comes great responsibility” and the same now applies for data scientists. With the growing demand, it is only natural that there are now higher expectations for trained professionals who achieve success from the get-go.
According to Glassdoor’s composite data scientist job description, the minimum requirement is at least a master’s degree and five years’ experience in a data or statistical field—meaning that only the best of the bunch will be needed as time goes by.
4. Better opportunities and scrutiny in the private-sector
According to BLS data, it shows that as of 2016, only 20% of data scientists worked at companies focused on computer system design and related services, and another 6% at software publishers.
However, 36% of data scientists were hired by government entities or higher education, and 17% worked in research or science-based environments—pointing to the fact that most private companies haven’t imbibed the data science culture.
Based on this data, it is expected that in the long term, once the private sector can fully get up to speed with data science and AI strategy, there would be numerous opportunities availing themselves to data scientists.
Liu noted that best organizations make optimum use of their data scientists.
“A healthy company with proper data science leadership will already make a distinction between data scientists and data engineers or data labellers and will often hire vendors to do data annotation work,” she said.
5. Bigger strategic and social influence
Due to the growing importance of data science to the well-being of an organization, it will give data scientist the upper hand over the direction an organization takes.
Rather than being relegated to side conversations, every data science team would now be granted the influence to remain in constant conversations with key stakeholders in the business.
“Data scientists are expanding their roles beyond finding solutions to predefined problems,” said Liu. “They’re also identifying opportunities where data science can help organizations grow their business. Business leaders in many organizations today are working with data science leaders closely to refresh and update their strategy.”