One major complaint we hear from companies working with Big Data is that their data scientist is also their data janitor — in other words, he or she is wearing too many hats. All of this multitasking deprives the data scientist of time to drive profit with real innovation.

A second problem some businesses face is essentially the opposite: They have several data employees all doing different tasks, and it’s like herding cats getting them to synthesize their work into practical solutions. So it seems whether your data staff consists of one or many, you can’t win with the current operation models available.

Bespoke Big Data

Dinesh Nirmal, VP of Development for Next-Gen Analytics, Spark and Hadoop at IBM, says companies with multiple data “personas” need choices, convergence and simplicity in their strategy. “How do we make sure that all these personas can collaborate between each other?” he asked.

He told Dave Vellante (@dvellante) and Jeff Frick (@JeffFrick), cohosts of theCUBE, from the SiliconANGLE Media team, during BigDataNYC 2016 that IBM’s new offering DataWorks allows teams to choose among tools like R, Python and Spark, “so you have a variety of choices as a data scientist to use the different algorithms and execution engines that we provide.”

They also have choice in how data models and algorithms are deployed: Batch, streaming or real time.

No data scientist? No problem

Addressing the other end of the spectrum, Nirmal said even an organization with zero data scientists can use DataWorks for end-to-end solutions.

“What it does is you throw the dataset at it — it will pick the best algorithms, which makes your job a lot easier,” he said.

Watch the complete video interview below, and be sure to check out more of SiliconANGLE and theCUBE’s coverage of BigDataNYC 2016.

Photo by SiliconANGLE