Applying for a data scientist job can be an intimidating task as there can be many things to take care in an interview process — right from justifying the practical knowledge to showcasing the coding skills. While we have earlier discussed articles on how to crack data science interview and what are the things to keep in mind while appearing for an interview for data science-related roles. This article deals with some of the things that you might be doing wrong if ever you are rejected in a data science interview.
Here are five things you may have been doing wrong:
Not focusing on the job description: The definition of data science jobs is not always the same and may mean different roles and responsibilities for different companies. Some of the commonly required skills may be a PhD in statistics, Excel skills, machine learning generalist, Hadoop skills, Spark skills, among others. The job description largely varies for every company and it is important to thoroughly dig it and carefully look for specific skills, tools and languages. It is important to display the skills that the potential recruiter is looking for so that they can shortlist you easily.
No specific distinction of technical skills: The technical skills in data science and analytics industry is quite wide and not mentioning your strengths correctly might jeopardise your chances of cracking the interview. For instance, it might not be apt to just say machine learning skills as it might include a whole spectrum of things ranging from linear regression to neural networks. And these sub-areas might further require knowledge of specific tools and software such as Python, Keras, R or Pandas. It is always advisable to give specific skills that you master than describing generic skills as might confuse recruiters of the exact skills that you pose.
Incorrect information and rephrasing work experience: To suit the data science job roles, many a times candidates rephrase their previous work experiences such as in the IT or software domains to present it as data science job roles, which might disguise your abilities initially but expose the depth and understanding of the skills later. You might have included job description aligning in a way that suits data science job roles but you might not have a deeper experience in it, which may get noticeable by recruiters during a one-to-one interaction. Mentioning of incorrect or misleading facts may also lead to recruiters rejecting you. For instance, the resume may state achieved an accuracy of say 90% on the test run, but what are the baseline and state-of-the-art score for this dataset to claim these numbers?
No mention about the projects that you have worked on from the scratch: Many times the only projects that a candidate mention in a resume are the ones they have done on Kaggle. While Kaggle is a platform for a lot of researchers to explore avenues in data science, it also serves as a source of practice for people who aren’t a pro in data science field and are trying to make a transition, mentions a recruiter in one of the forums. There are different kinds of the audience at Kaggle such as those who are playing around with the dataset or getting to know how problem-solving in data science works like, without having actual experience in solving or creating a new data science problems. So, listing just Kaggle project might be good but not definitive of how good your data science skills are. Even if it a Kaggle project, it is better if it is done from scratch. Other than that, it is important to mention the projects that you have worked on. It gives recruiters a chance to understand the problems you faced and the way you approached the problem, thereby giving them a glance at your problem-solving abilities.
The resume is full of buzzwords and no concrete proof of your skills: While the resume may suit the job description, but there are chances that you are rejected if there are too many buzzwords in the resume and no concrete way to prove that you actually pose those skills. You may mention in the resume that you have had experience with Hadoop, Excel or certain areas, but if you have showcased it real-time on platforms such as GitHub, it convinces the potential employers of the skills you have. They can look through various projects you have been a part of and see how you have dealt with real data. Hiring managers like to see the time that a candidate has spent from start to finish. Having a portfolio gives recruiters just that. There may be fancy sounding terms in the resume, but if you don’t have a proof to showcase it, you might be rejected for a potential data science job role.