Following up on his popular interview, we have caught up again with Vikas Grover, Senior Manager in the Data Science chapter in Vista, to discuss what makes a great Data Scientist!
-
Currently working as a Senior Manager in the Data Science chapter in Vista, I have amassed over 10 years of experience in this dynamic and rapidly evolving field. From the early days when the term ‘Data Science’ was shrouded in mystery, to the present era of ‘ChatGPT’ and beyond, it has been my privilege to continue to be part of this fascinating journey.
Despite the emergence of advanced AI-powered tools and the maturation of the industry as a whole, I believe that the key to success in Data Science remains grounded in a deep understanding of the essential skills required for the job, coupled with an unwavering focus on solving real-world customer problems. Here are my top 5 skills to succeed in Data Science.
- Fundamentals of Data Science: Nowadays, building predictive models that work is a ‘piece of cake’ thanks to AutoML and the abundance of reference material available on the internet. However, producing high-performing models that solve business problems requires some solid skills. To master this craft, you must devote time to learning various machine learning concepts, such as supervised learning vs unsupervised learning, regression vs classification, cost functions, regularization, underfitting and overfitting, evaluation metrics, cross-validation, etc.
For me, Andrew Ng’s courses on Coursera provided the most clarity on basic concepts. I personally recommend them. - Programming: In order to transform raw data into actionable insights and decisions, a data scientist must have advanced programming skills. Today, Python is considered the preferred language for programming due to its capability to handle every aspect of data science processes, including the deployment of models. It is expected that an expert data scientist is also an expert programmer. Here are some courses I enjoyed in the past:
a. Python for Everybody
b. Crash course on Python - Maths and Statistics: Data Science requires a solid grasp of maths and statistics since these disciplines are the basis for all machine learning algorithms. Business owners, especially those that are data-driven, will expect Data Scientists to understand different approaches to statistics in order to make recommendations and decisions. A deep understanding of the mathematics behind the optimization of cost functions of popular machine learning techniques could help one stand out in the crowd. Have a look at:
a. Mathematics for Machine Learning
b. Statistics for Data Science and Business Analysis - Business Acumen: A good data scientist must have technical skills, but an effective one must have business acumen as well. It is essential for a data scientist to fully understand the business problem and its key goals before jumping into a data science solution. The best solution may not always be an ML model, and an expert data scientist can explain ‘why’ based on business context.
- Communication: Often overlooked in technical roles, communication and storytelling are key elements when it comes to the successful implementation of a data science project. Effective communication is needed throughout the process, whether explaining your approach and solution to the team, and stakeholders or presenting to business leaders. One needs to be able to communicate with difference audiences and at different levels of seniority to be able to deliver powerful ideas and ultimately advance in their career.
Thinking of making a career leap into a data-driven organization? Explore possible career opportunities with Vista DnA.