Data scientists are the most wanted professionals in various industries with companies and startups looking for a qualified data scientist to join their teams. While we have covered a lot of articles around what a data scientist job is all about, things a data scientist should have in their resume must have skills for data scientists and others, we bring an exclusive industry view on what it takes to be a good data scientist.
For this month’s theme, we spoke to various data science heads in the industry who have shared their perspective on what it takes to be a good data scientist. The series of interviews would be especially helpful for those looking out to transition into data science team of the companies.
Our first interaction is with Dr Sunil Bhardwaj, who is the currently working as a senior analytics training consultant with SAS India. He is a SAS Certified Data Scientist and has does various partner enablement for various ongoing projects for partners such as KPMG, Wipro in Antimony Laundering, Credit Risk and related areas. He has more than 15 publications in the area of Analytics and Quantitative Techniques in national and international journals.
Analytics India Magazine: What are the key skill sets that you look for while hiring for data science roles, in terms of languages and technical skills?
Sunil Bhardwaj: The key skill sets required for a data scientist role include:
- Data Management including Big Data, Unstructured data (Knowledge of HDFS, NO-QL, HIVE, PIG, SAS etc.)
- Analytics including AI, Machine Learning, Forecasting, Optimisation and Experimentation (SAS, SAS Viya, R & Python)
- Data Visualisation (SAS Visual Analytics and other visualisation tools)
AIM: What are the non-technical skills and traits that a good data scientist should have?
SB: Technical skill helps with hands-on experience but the ability to correctly interpret & communicate this to a general audience becomes more important. Communication is the most important skill that translates analytical findings to a non-technical audience. This skill ensures the final usage of analytics and data science work to be consumed or utilised for the overall business benefit. Three important traits that are important in my opinion from this perspective are:
- Team Player
AIM: Do you believe that a good data scientist should be obsessed with solving problems and not new tools?
SB: Solving problems is fundamental to a data scientist. However, tools are equally important. Imagine a surgeon without his surgery kit! Tools are the enabler for any data scientist. Better and faster the tools, better the delivery and solution.
AIM: Is it educational qualifications or experience that matters more to be a data scientist in companies?
SB: Basic educational background matters a lot. Educational background with analytical, mathematical, statistical, basic sciences and computer science courses or skills are pretty fundamental and have an advantage over other traditional courses.
AIM: Who would be a preferred candidate for data science role — one with certification in a full-time course or the one with the executive course?
SB: It depends on the role and the job description. A suitable interview process is still required to gauge the knowledge and skills of the individual.
AIM: What is the best learning curve for a data scientist and the best resources to learn?
SB: Typically, a good data science course may have a duration from 6 months to 2 years. The learning typically starts with fundamental data management and statistical topics and ends in contemporary concepts of machine learning and artificial intelligence. Today there are many resources to learn and become a data scientist. These are:
- Full-time and Part-time University courses (ISB, IIML, IIMB)
- Full-time Classroom or Self-paced e-learning Corporate courses (SAS academy of data science)
- MOOCs (Coursera)
Some of the subjects a budding data scientist should strengthen their base are: Mathematics & Statistics, Computer Science (SQL and OOPS), Big data (Hadoop, HDFS, HIVE, PIG)
AIM: What is the importance of industry mentors for a budding data scientist?
SB: Industry mentors encourage & guide the career path to success. They are highly valuable counsellors who inspire the next-gen with their experience and knowledge. They can provide a listening ear to mentees, keep a watch on their goals, and challenge their thinking to bring out the best in them.
Mentees can learn from their mentors:
- Practical insights
- Real life case studies
- Real challenges for a Data Scientists
- Mentoring for the job market
AIM: How important is the knowledge of the sector for being a good data scientist?
SB: Domain or sector knowledge is important. Each domain requires a different knowledge type. The healthcare domain may require a different type of knowledge as compared to a financial market domain. The higher the domain expertise or knowledge, the better it is. The minimum requirement, a data scientist should possess is: understand what exactly the sources of the data are, how the data was collected, what biases may exist in the sample, what do the variables represent and what are the objectives of modelling in a domain.
AIM: In a nutshell, what are the 3 must-have skills for a data scientist?
SB: The three must have skills to be a data scientist include: Data Management, Data Modelling or Analytics, Data Visualisation along with business communication.