With data science having captured the imagination of job-seekers across a spectrum of industries, it may be interesting to pull out insights from the experience of seasoned data scientists on what aspirants should expect when they begin to look out for jobs.
Whether they are newly-graduated or looking for a career-switch, data science aspirants can take valuable lessons from those on the field when it comes to breaking into this industry and launching a career here.
Sign up for your weekly dose of what's up in emerging technology.
If our interactions with practising data scientists are any indication, collecting certifications and building knowledge without context will only take you so far. Securing a stable job in data science can be challenging, despite surplus employment opportunities available today.
One of the main reasons is because job interviews typically follow a rigorous format that could involve presentations and assessments as well. Even top companies like LinkedIn and GoJek evaluate other skills and do not prioritise formal educational qualifications alone when hiring for these positions. With the objective of providing data science aspirants with a comprehensive view for setting their expectations when scouring for jobs, we spoke to four data scientists to understand this better.
All unanimously agreed that the experience — although challenging — has been enriching as well. Moreover, most accept that mistakes will certainly be part of the journey for all — key will be to introspect, pivot and try again.
Importance Of In-Depth Knowledge Of Key Concepts
Considering the depth of technical knowledge required for any data-focused role, there is no feigning the level of expertise you will need to have to even pique the interest of recruiters.
“The primary qualities that organisations look for in data scientists is a proper understanding of statistics and sound logic. If you cannot demonstrate that, you may set yourself up for disappointment,” says Abhishek Goel, Head of Data and Analytics at To The New.
To develop a more nuanced understanding of these concepts, it is imperative that you are motivated by more than the prospect of a cushy job. Your primary driver should always be a passion for using data to solve practical problems.
Concurrently, you should not ignore the fundamentals when looking to amplify your knowledge.
“It is a challenge to find a data scientist who appreciates the working of algorithms, understands the math behind the algorithm, and is well-versed with the fundamentals as well,” says AI and ML specialist at Publicis Sapient, Sray Agarwal.
Concurs Dhartish Khatri, a data practitioner who found himself in a fix when he was unable to articulate a very basic concept during an interview.
“Oftentimes, when you prepare for an interview, you go very deep into the concepts but fail to articulate the basics. Do not make that mistake,” he says.
Significance Of Nuanced Understanding Of Business Context
While we explored the importance of being proficient in both fundamental and advanced knowledge of data science concepts, awareness of the industry the company operates in is crucial as well and is likely to be evaluated during the hiring process.
Explains Goel, “With the advent of technologies, a lot of frameworks and API are available for data scientists to use. Many data scientists apply these without understanding the behind-the-scenes logic. So what essentially happens is that everything is black-boxed. As a result, data scientists often provide an output, without understanding the reasoning behind it.”
This problem could largely be mitigated if candidates understand the sector well, as well as the problems the company they have applied to is working to solve. Armed with this knowledge, they may successfully be able to extract meaningful insights and make useful recommendations — skills that are always put to test in interviews.
Unfortunately, this aspect has often been ignored by data enthusiasts.
“It is rare to find a data scientist who views a problem from a business and problem-solving perspective, rather than just applying random ML algorithms,” says Agarwal.
Even here, what is explored within the realm of an interview may still offer a fraction of what you may have to deal with in the real world.
Agrees Goel, “Data science problems in real-world tend to be open-ended, unlike the closed-loop problems with defined constraints that aspiring data scientists may have to solve in interviews and competitions.”
Why You Should Be Active On Platforms Like Kaggle
Investing time and resources into creating an impression on online platforms like Kaggle may be worthwhile — an endeavour that Prashant Kikani, a data scientist at an ed-tech startup found a lot of value in.
“I started dabbling with Kaggle over two years ago with the sole purpose of updating myself with the latest tricks and tips in the data science industry,” says Kikani. “Not only does it help in building your knowledge but also provides opportunities to earn titles that could potentially get you noticed by the right people,” he adds.
Kikani claims to have landed a lot of job interview offers through Kaggle on the back of his projects.
According to him, Kaggle also offers users the opportunity to see codes written by other people via their kernels and allows them to participate in discussions with like-minded people — an exercise that has enriched the job interviews he has given.
Moreover, he strongly believes that it could be a good place for aspirants to kick start their journey in data science since it covers a wide gamut of skills required, including data wrangling, pre-processing, and building a whole end-to-end pipeline.
“There is a very good culture of sharing and building knowledge in Kaggle as well — people will encourage you if you are creating good content,” adds Kikani. However, he warns against taking titles too seriously. “Core focus should never be on earning titles — they are just a means for you to take advantage of a bigger opportunity,” he says.