Home » Students Blog » My early days in the data science industry…

My early days in the data science industry…

‘Data science is a vast field that cannot be summarized with a few words. Although we could simplify it by saying “anything that involves data.” Generally, any data science course curriculum tries to cover every sub-field, such as Artificial Intelligence(AI), Machine Learning(ML), Deep Learning(DL), Data protection, etc., but in the corporate world, only a few of those subfields come in handy for daily use. For instance, in my role as an executive data scientist at NielsenIQ, I have been dealing mostly with python and statistics, although the concepts of cyber security, data protection policies, basics of data analysis, Machine Learning, and Deep Learning have come in use once or twice. This doesn’t mean that these concepts aren’t useful in the company. There is a cartography team that deals with geospatial aspects using machine learning algorithms. All the subfields of data science are required in the industry, but it isn’t expected to work using all. The workload is divided into teams, and only a specialization of the set of sub-fields of data science is assigned to a particular job role. So it is important for academic institutions to teach data science concepts and, at the same time, provide a platform to students to specialize and focus on a certain set of sub-fields according to their interests. This could be helpful for them in cracking their interview to get their desired job role.

Coming towards the non-technical part, in my opinion, “learning from professors” is important, but “learning from peers” is also as important. In the corporate world, there is no professor-student-like system. It’s an individual’s responsibility to ask around colleagues, learn and ask as many questions as possible while doing work. I believe academic institutions should create such an environment where “peer-to-peer learning ” happens that would give them practice on how things happen in offices.

Last but not least, knowing how to interpret output is significant. Even if the model is efficient and accuracy is excellent, if you cannot deliver or understand the significance of obtained results, the whole project is meaningless.

Kashish Gajwani
Student, M.Sc(Data Science & Spatial Analytics)
Batch: 2021-2023

 

Leave a Reply

Your email address will not be published. Required fields are marked *

Name *
Email *
Website