Exclusive Interview: Ajay Bhargava, TCS on the Ideal Analytics Curriculum at Graduate-level
We discuss the differences between analytics and Big Data, the evolution of expectations from data science, important qualities desired in data scientists, ideal curriculum for Analytics focused programs, advice and more.
Ajay Bhargava has more than 25 years of industry, research, strategic consulting, and teaching experience in the areas relating to databases, enterprise data management, data warehousing, business intelligence, advanced analytics, and Big Data. He has also contributed to SQL, ODBC and IDAPI database standards. Ajay built, and heads up the Global Analytics & Big Data practice for TCS Insurance & Healthcare customers.
He has frequently spoken at industry conferences, authored whitepapers, and has driven thought leadership in the Data industry. In addition, he has actively taught (Analytics, Database Design, and Data Mining etc.) at The University of Texas, Austin, and College of Engineering, Pune, and mentors high school entrepreneurs for global competitions.
Ajay holds an M.S. in Computer Science and M.S. in Aerospace Engineering from The University of Texas at Arlington. He obtained his B.Tech in Aeronautical Engineering from Indian Institute of Technology, Mumbai in 1984.
First part of interview.
Here is second part of my interview with him:
Anmol Rajpurohit: Q5. What is the importance of having a Data Governance model? What approach do you recommend for setting up Data Governance model?
Ajay Bhargava: Having a Data Governance program increases the odds of success in any data initiative. The program must be established at a strategic, tactical, and operational level. The program must encompass people, process, technology, and data dimensions of the program. The involvement of both IT and Business is crucial for a timely, high-quality, within budget delivery of initiatives.
The key is to plan and architect and design for the future, but implement and execute in the near term, starting with small initiatives. Key is to continuously deliver value to business, while seeding an analytics-driven culture in every aspects of business. A strong data governance program provides a rich career path for data science teams, (data, tools, and analysis) synergies across different lines of business, and allows IT to deliver on multiple, sometimes competing objectives by various business initiatives.
The key is to plan and architect and design for the future, but implement and execute in the near term, starting with small initiatives. Key is to continuously deliver value to business, while seeding an analytics-driven culture in every aspects of business. A strong data governance program provides a rich career path for data science teams, (data, tools, and analysis) synergies across different lines of business, and allows IT to deliver on multiple, sometimes competing objectives by various business initiatives.AR: Q6. What are the key differences between Analytics and Big Data? How can one assess if the solution requires Analytics or Big Data or both (or none)?
AB: As explained earlier, big data’s “harvesting” aspect is achieved using analytics. But, irrespective of a big data platform (harnessing) or not, you need to use analytics to derive actionable insights and value. If the solution demands leveraging and investing in a big data platform, to deliver value when you want to
- Achieve business SLAs that are not possible to be met by traditional environment e.g. speeding up ETL in a data warehouse environment
- Introduce new capabilities, such as leveraging unstructured data for fraud detection, sentiment analysis, real-time triaging of tweets, and enterprise search across intranet and internet etc.
Most of these initiatives start as a proof of concept. Many a times, a big data platform is introduced (as a proof of technology), along with specific use cases (proof of value) to lay the foundation and enable value creation for the future.
AR: Q7. How do you think the expectations from Data Science have evolved over time? Where do you see them headed in the future?
AB: In the eighties and nineties, the focus was more on decision support systems to make better informed decisions from structured data in relational databases and mainframe systems. The emphasis was more on computing quicker and less on the analysis of data.
In more recent years, the shift has been more towards analysis from not only structured, but unstructured text, audio, and video data. The revival of machine learning techniques from artificial intelligence days, the application of cognitive computing, along with advances in natural language and speech processing, massively parallel computing environments, analyzing large amounts of generated data, more and more personalized and prescriptive analytics, coupled with deep focused knowledge of business domain problems is where I believe the expectations from data science are heading.
AR: Q8. What key qualities do you look for when interviewing for Data Science related positions on your team?
AB: To me, a person’s attitude, willingness to learn, and analytical techniques used to improve one’s learning are pretty important to look for in a candidate. In today’s world,
having a diverse group of individuals from a multi-disciplinary field brings balance and perspective to solving business problems by a data science team. The science of problem solving (the “HOW”), art of storytelling from what the numbers and visualization are trying to tell you, the perpetual intellectual curiosity of asking the “WHY” behind the “WHAT”, listening to needs of business etc. are some examples of qualities I constantly try to gauge, when interviewing for data science positions in my team.
AR: Q9. Based on your experience of Analytics field from industry as well as academia perspective, what do you believe should be the focus areas of Analytics education in universities in order to prepare students for the real-life Analytics projects?
AB: There are quite a few skills that need to come together for real-life analytics initiatives, not all of which will be present in any one person. There are numerous graduate level programs that have started in various universities over the last five years or so, to try and fill the enormous gap between low supply and high demand for analytics professionals worldwide.
I believe that a 2-year graduate program that not only dives into data management, applied statistical and machine learning techniques, practicum with real world industry data, but also has some elective courses that are industry domain specific added to the mix, is required to prepare students for real world customer issues. A real symbiotic, partnership (to create a pipeline) between academia and industry is absolutely essential.
In addition, introducing concepts and career options, and spotting and grooming such talents as early as high school level, is a must for any company, industry, or for that matter, country to maintain a competitive edge.
AR: Q10. What is the best advice you have got in your career?
AB: I once read somewhere:
“Pursue and excel in what you love, and get someone else to pay you for it” :-)
AR: Q11. What was the last book that you read and liked? What do you like to do when you are not working?
AB: I recently read “Calculus – Early Transcendentals” by James Stewart. It explains calculus from multiple angles. Feel like taking calculus all over again from the eyes of this textbook.
When I am not working, I like to travel with my family and meet folks from different cultures, and when possible, watch some college football and international tennis.
Related:
