Vishal Gupta is an alum from the 2017 batch, who graduated from the department of Chemical Engineering with a BTech degree. Currently he is pursuing MS in Business Analytics at the University of Texas at Austin. In this blog he talks about how he navigated a career in the domain of Analytics. Read on to learn more about his journey during college and the valuable insights he has for anyone planning to pursue a career in Analytics.
Before the job:
- When did you decide (in which year or a specific event) to go for Analytics?
During my second and third year, I sort of started exploring a lot of different things: core industry through industrial visits and attending core talks, research through projects and university intern, software development – another intern and finally analytics through Coursera and coursework at IITB. I was not sure till the start of the placement semester what I wanted to go for, but from whatever information and experience I gained from doing different things, I concluded that analytics is more suited for me and something that I can see myself working long term in the future.
- How did you build up your skill-set in the field? e.g. activities on campus, internships, or courses inside or outside the institute
I have built up my skills majorly from doing courses – both inside and outside the institute
- Coursera – Data science specialization
- Coursework – Intro to data analysis, Applied Multivariate statistics (chemical dept. elective), Probability and Random process (electrical minor)
- What were some of the essential resources you referred to for your preparation?
The resources back then were very limited but nowadays you can find tons of articles on the internet, although they can be confusing. So even to this date, I try to stick to limited but reliable sources. What I specifically used during my preparation were
- Books – Introduction to statistical learning and Elements of statistical learning, both are available online for free
- Blog – https://www.kdnuggets.com/,
- Notes from coursework and online courses
On top of these, there are two more blogs that I wasn’t aware of during my placement but now consistently refer to learn something new –
- How important were each of these factors in helping you secure the job: CPI, PORs, soft skills, Coding practice for particular coding languages (Python, R, SQL, etc.)?
- CPI – there was a cut-off of 7.5, for some companies it was even lower (7), but beyond that, it didn’t matter that much. Most of the campus hires in analytics come from not-so-relevant domains (in my case – chemical engineering), so having decent academics is a good enough indicator for your performance once you are hired.
- PORs – I didn’t have any third-year POR. It’s good to have something to talk about but wasn’t necessarily required.
- Soft skills – very important; usually I was asked to explain a machine learning model (linear regression almost all the time) or asked to solve a case that uses one. So it’s very important to convey the nitty-gritty of the algorithm as well as how you can leverage the algorithm to solve a business case.
- Python or R – being proficient in any one of those languages is a must, there were Python-related questions in the company’s tests and I was asked to solve a basic python question during the interview itself
- SQL – I knew SQL beforehand, which was a plus, but I wasn’t quizzed on my SQL knowledge during the entire process any question on SQL
- If you were to start all over again, what would be your roadmap to analytics?
I will not change anything that I had already done, but would want to add a few more things on top to make myself a more suitable candidate –
- Kaggle competitions or independent analysis on data
- Internship and projects – from coursework, online courses, or under a professor
- Git repository/page – to showcase the work done – codes, visualizations, etc
- Interview practice – interviewers vary significantly during analytics interviews, while some are very technical, some can be very business-oriented. One should be able to understand when to dwell into the depth and when to cover the breadth
- According to you, how different is the preparation for the analytics role from Software development and other IT roles? Is there any significant overlap (Even in the industry)?
The preparation for analytics is very different from software development and other IT roles. While the former requires basic programming (Python or R) and statistics, the latter requires object-oriented programming, different languages, and technologies. Even in the industry, there is a significant difference in the tools that each utilizes. So if one is starting the preparation from scratch during the placement semester itself I wouldn’t recommend going for both roles simultaneously.
- What made you finalize Business intelligence Analyst at Target? What aspects of the company/career path should one consider while making this decision?
Target is a US retailer, with Walmart and Amazon as its competitors. It had in-house data science and analytics working on complex problems. On top of that, they leveraged Big Data tools like Hadoop, Spark, Hive, etc. which is something that wasn’t being used by most companies back then. Based on this information, I knew I would get ample opportunity to learn various tools and technology that Target has at its disposal and my work will have a direct impact on the organization.
During the Job:
- What are the hard/soft skills essential in your job? What were the new skills you had to acquire later in your career but probably not essential during the placement?
There are a ton of hard skills that can be used in an analytics and data science project, and any combinations of them are used based on the need. Exposure to most of them is almost impossible for a freshly graduate student. To name a few tools and technologies –
- Hadoop, Spark
- PySpark, SparkR
- AWS stack – EC2, S3, SageMaker etc
- Visualisation tools – Domo, Tableau, Power BI
- Containerization – Docker
- CICD – Jenkins, Drone
I learned most of these while working on various projects, these are some of the things that you leverage to build and productionalize a solution. Oftentimes it’s not necessary to understand the ins and outs of them. All of these skills are relevant when you are working in an organization, but as a student trying to make a career in analytics, Python/R and stats are more than enough.
As important as they are, one can’t simply ignore story-telling. As an analyst, when a business problem is posed to you, you have to decide what tools or technologies to use but the adoption of your solution and insights hinged a lot on your ability to tell a cohesive story and make people understand why you did, what you did and what impact it can have on business. Sometimes it does happen that the analysis or the solution is rejected as it is incomprehensible to people who posed the problem to you.
- Could you please tell us more about the company culture and Work-Life balance in the Analytics sector (Could be in general or specific to the company)?
Culture and work-life vary a lot across different organizations. Since analytics is somewhat domain-specific it varies from – retail to fintech, consultancy to in-house, start-up to MNCs. I have worked in two companies so far – the first one a Fortune 50 company and another retail pharmacy start-up based out of Mumbai. The work-life balance and culture are polar opposite. Work hours were almost fixed except for few calls a week with US counterparts,
- If someone wants to switch to a managerial role, how easy or difficult is it for him/her to make this transition after starting a career in analytics?
Managers in analytics and data science are with at least 5-8 years of work experience, so making a transition from individual contributor to people’s manager is very difficult. Instead of transitioning vertically, what people usually prefer is to change the role itself. Analytics in itself is a very broad field that has a lot of different roles based on the rigour needed –
- Business analyst, product analyst – more business and product-oriented
- Data scientist, machine learning engineer – deals with building complex algorithms and productionalising the solution
- Data engineer – creates data pipelines and data warehouses for the others to consume
You might have landed a specific role during placements and after, let’s say a year or two, want to switch to a more engineering profile or business profile. These kinds of transitions are relatively easier, based on personal interest and skillsets, and are seen quite often within organizations or shifting from one to another.
- What’s next in store for your career/ job profile? Have you given a thought to any future plans?
I am currently pursuing MS in Business Analytics at the University of Texas at Austin. Over time, I have realized you can learn only so much working on the job and how essential it is to have structured learning if you want to deepen your technical knowledge
- What impact will the COVID-19 pandemic have on your job profile, your current employer, and the sector as a whole?
Covid has significantly affected the job role, since collaboration is a key part of what you do as an analyst, either with your team or stakeholder. Going online has made things a lot more difficult. What can be an easy 10 min break conversation has now turned into an hour slot meeting, which often takes 2-3 days to schedule. The number of online meetings has significantly increased, sometimes even taking the major part of the day. Informal brainstorming sessions are completely gone. In a start-up specifically, where the primary source of information is informal chit-chats and tea breaks, information flow gets restricted and often delayed as no formal channel is established.
A word of advice to the people sitting for placements this year?
I would recommend not stressing about getting a specific job or company; it’s just your first job. It may seem that if you don’t do well it might affect your career but that is not the case. Over the years I have seen a lot of people who didn’t start well or didn’t get the company or profile they want during the placements, but have worked their way and got a great profile later in their career. A lot of us during placements are not even sure as to what we want to do. Wherever you land during placements, it will help you in figuring out what you want to do in your life or what you don’t.