Now Reading
The 3 Ps To Be A Good Data Scientist Are Programming Foundation, Passion And Patience

The 3 Ps To Be A Good Data Scientist Are Programming Foundation, Passion And Patience

Srishti Deoras
Emudhra Banner

Emudhra Banner

For the next interaction in our series of ‘How to be a good Data Scientist’, AIM caught up with Kaushik Srinivasan, Senior Vice President of strategy and Innovation at eMudhra. Srinivasan is one of the founding members of the company, a licensed Certifying Authority (CA) of India issuing digital signature certificates. The company is at the forefront of driving paperless transformation as a part of Digital India campaign and also adopts recent technological innovations such as data analytics and blockchain in its daily working. With a team comprising of Data Scientists and AI researchers, hiring a good data scientist is crucial for driving success and innovation. In a freewheeling chat, Srinivasan shares his views on what it takes to be a good data scientist.

Analytics India Magazine: What are the key skill sets that you look for while hiring for data science roles? What are the languages and technical skills they should know?

Kaushik Srinivasan: Data Science is a broad area and there are a number of tools and languages that are emerging and are promising and when it comes to crunching large volumes of data. We look for knowledge in Python, Scala and more recently the ability to work on Apache Spark and Tensorflow.

AIM: What are the non-technical skills and traits that a good data scientist should have? How important is effective communication and business mindset for being a good data scientist?

KS: The ability to translate business to technology is critical for a data scientist. If you are trying to understand the customer churn at a Bank v/s an Insurance company; while the tech stack that you will use is more or less the same, the domain understanding and the ability to map that to the input you need to use and ability to then select the right algorithm play a vital role in getting desired outcomes. If this is not done correctly, there is a high likelihood of getting false positives. The other quality required is patience since accuracy in data science is a function of iterative trial and error to get the right outcome.

AIM: Do you believe that a good data scientist should be obsessed with solving problems and not new tools?

KS: I would say it’s a combination of domain and technical understanding which is why the role is valued so much in the industry today. Today most complex problems including health diagnosis and treatments are done through technology, more specifically artificial intelligence. Each tool or platform has its own strengths and weaknesses. So, unless you are aware of which tool/platform is most suited to solve the specific problem, there will be a lot of unnecessary effort put in the wrong direction.

AIM: Is it educational qualifications or experience that matters more to be a data scientist in companies?

KS: Formal education in machine learning has more or less become a major prerequisite today. I think that someone with a lot of passion can substitute this requirement but developing a knack for problem solving through real world experience goes a long way in developing a person’s caliber, as they are better suited in their ability to solve complex problems.

AIM: Who would be a preferred candidate for data science role—one with certification in full time course or the one with executive course?

KS: As stated in the previous answer, we look for people who are passionate and who have a background in technology, and we train them while they are on the job. A good certification is certainly desirable but in no way it is an eliminating factor.

AIM: What is the best learning curve for a data scientist and the best resources to learn?

KS: The best learning will be on the job as the data scientist would be looking at data first hand and understands the use of data to drive decision making. Given today’s systems and computational power, data crunching has become a real time exercise thus giving data scientists who have the passion, a quick ability to impact the company’s bottom line.

AIM: What are the subjects a budding data scientist can master during the early days of his/her education and career to be a good data scientist?

KS: Strong understand of programming concepts such as object oriented programming, database design, clustering, multithreading are a must.

AIM: What is the importance of industry mentors for a budding data scientist?

KS: Industry mentors bridge the business technology gap, and getting a good mentor will always help accelerate the learning curve of a budding data scientist.

AIM: How important is the domain knowledge for a data scientist?

KS: Domain understanding is critical to succeed in the data Analytics world, but then again this is something that has to be learnt on the job.

Provide your comments below


Copyright Analytics India Magazine Pvt Ltd

Scroll To Top