SAN FRANCISCO — Data scientists have been labeled the new “rock stars” by analysts, and demand for their talents is heating up from startups and large corporations.
Technology consulting firm McKinsey is already forecasting the US will face a shortage of up to 190,000 data scientists by 2018. To fill in the gap, education technology companies like Coursera and Udacity are producing out online tutorials to help anyone master the basics of data science. In addition, companies are developing easy tools for developers and analysts to churn out data visualizations.
At DevBeat, our developer conference Hulya Farinas provided some rare insight into how her company Pivotal is developing tools to make its customers’ applications much faster. Farinas is a data scientist, who previously worked at IT giants like Greenplum and IBM.
Health care is the prime sector for Pivotal, a cloud and analytics company that is already working with hospital networks and insurance providers like Kaiser Permanente.
On stage, Farinas described how Kaiser recently invited IBM, Pivotal, Cloudera and Hortonworks to a unique hackathon. These rivals were asked to crunch Kaiser’s data sets, and determine the level of correlation between air quality and demand for medication.
With a more precise understanding of how weather and air quality is linked to asthma, Kaiser can better prepare for an influx of patients who will require medication. According to Farinas, Pivotal was able to perform complex this data analysis and statistical modeling in seconds. “We determined the zip codes where asthma are overrepresented,” she said.
This example is the tip of the iceberg for health care — in future, the explosion of wearables, sensors and remote patient monitoring devices will mean that health providers will have far more data to work with. Our Fitbits or Jawbone Ups are designed to be worn every day, so this data won’t just help us learn more about sick patients — but healthy ones too.
Indeed, the demand for data scientists like Farinas is escalating, and salaries are ramping up. Increasingly, we’ll see developers seek to build up their skill-sets, and learn statistical languages like R, in order to land prestigious data science jobs.
Farina’s advice to budding developers who are interested to learn more about data science? Begin with resources like “a programmer’s guide to data mining,” and strike up conversations with established data scientists. “It’s a very tiny jump,” she said.
VB's research team is studying web-personalization... Chime in here, and we’ll share the results.