Diving Deep Into The World Of Data Science With Ashutosh Kumar

Hire IQ by HackerEarth is a new initiative in which we speak with recruiters, talent acquisition managers, and hiring managers from across the globe, and ask them pertinent questions on the issues that ail the tech recruiting world.

Next up in this edition is Ashutosh Kumar, Director of Data Science, at Epsilon India.

We had a long chat about hiring for niche roles like data science and data analysts, whether there will still be a need for such roles post this layoff phase, and expert tips that developers can make use of to excel in these roles.

Dive in!

P.S. If you missed the previous edition of HireIQ where we sat down with Patricia Gatlin, Diversity Lead/Talent Sourcing Specialist, at Johns Hopkins, you can read it here 🙂

Let’s delve into the future of data science

HackerEarth: Can you give us a small bio of your journey in tech recruitment?

Ashutosh: I have been a part of recruitment in the data science field for nearly 14 years of my career and have recruited for successful startups (seed to Series D) and MNCs across levels (entry, junior, mid and senior management) and profiles including data analysts, data scientist, ML engineers, full stack developers, and DevOps/MLOps. I’ve also been part of campus recruitments in premier colleges (IITs, NITs, IIMs, and ISB) for roles in data science profiles, as well as the lateral hiring processes for experienced candidates for almost all my previous employers.

HackerEarth: In this era of mass layoffs, where do you see the data science and data analyst roles heading? Will there still be a need for this niche domain going forward?

Ashutosh: Mass layoffs depend on the health of a company and its measures to keep itself up and running and have less to do with any specific roles. Companies can cut all types of roles when it comes to survivability, but domains like data science and technology are some of the last ones to be axed since these are business-critical roles.

For instance, several of our clients, who are facing the pressures of recession, have been turning to data science to gather data-based insights on how to increase their revenue and save costs. Data science plays an important role in helping companies navigate and weather the recession storm.

We are a data-driven world, and data science will continue to be an in-demand domain. The demand for data science and data analysis professionals may fluctuate depending on economic conditions and the specific needs of individual organizations. It is important for professionals in these fields to stay up to date with the latest technologies and techniques, and to be proactive in seeking out new opportunities for growth and development.

Also read: Inside The Mind Of A Data Scientist

HackerEarth: What are some of the mistakes/misconceptions (top 3) that you have seen recruiters or engineering managers make when hiring data scientists/data analysts?

Ashutosh: Firstly, focusing only on interviews and theoretical questions instead of looking for hands-on coding experience is a big mistake. The industry needs people who can not only understand algorithms but who can also code. It’s fairly easy to get a theoretical understanding of all data science algorithms from the internet without writing a single line of code, and we need to ensure we hire people who can actually build solutions.

Secondly, giving importance to degrees and background over expertise. Today, there’s a plethora of online degrees which require little effort for a diploma or master’s degree in data science – one can get a degree from Indian or international colleges for ~USD 4000. Some of the best data science professionals we’ve worked with have unrelated degrees and have learned everything by themselves – either from online courses, Kaggle, blogs, or self-training.

Lastly, every data-related skill cannot be equated with data science and AI. The latter’s expanse is wide and complex – from simpler tasks like data entry, to intermediate ones like analysis, visualization, and insights, and to the more advanced machine learning models and AI algorithms. Often, roles are clubbed as ‘data scientist’ simply because of such loose definitions of these terms. You don’t need to hire a data scientist when you may actually need a data analyst.

HackerEarth: How do you see the new technologies like AI, ML, and quantum computing affect the field of data science?

Ashutosh: AI, machine learning, and quantum computing are all rapidly advancing technologies that have a significant impact on data science. AI and machine learning are enabling data scientists to develop more advanced algorithms and models that can analyze and interpret data more effectively, while quantum computing is providing the computing power necessary to process and analyze large amounts of data quickly and accurately. These technologies are also helping automate many of the tasks that were previously done manually, which is making data analysis more efficient and accessible. Overall, these new technologies are helping drive significant advances in the field of data science and are likely to continue to do so in the future.

Also, read: How AI Is Transforming The Talent Acquisition Process In Tech

HackerEarth: How would you recommend that data scientists upskill themselves to cope with the current and upcoming changes in the economy and the tech world?

Ashutosh: As a data scientist, it is important to continually upskill and stay current with the latest developments in the field. Here are a few ways data scientists can upskill themselves:

• Stay updated on the latest tools and technologies: Data science is a rapidly evolving field, and new tools and technologies are constantly being developed. There are new algorithms in the domain of Deep Learning, Reinforcement Learning, Transfer Learning, LightGBM, GANs, Transformers, large language models, and Explainable AI to name a few. There are new tools and frameworks in the industry like Airflow, Horovod, Petastorm, etc. developed by companies like Facebook and Uber, which have been made open source. There are also AutoML, ETL tools, visualization tools, cloud enablement tools, collaboration, and project management tools (like Asana and Trello). Keep abreast of these advancements and use them effectively in your work.
• Learn new programming languages and frameworks: As a data scientist, you’ll need to be proficient in at least one programming language, such as Python or R. It is also useful to learn additional languages and frameworks such as SQL, Julia, or TensorFlow.
• Enhance your machine learning skills: Machine learning is a key aspect of data science, and it’s important to have a strong foundation in this area. There are many online courses and resources available to help you learn machine learning and apply it to real-world problems.
• Stay informed about industry trends and developments: There are various forums on the internet that track the latest trends and developments in data science and machine learning. I follow researchers, data scientists, machine learning experts, and AI/ML companies on Twitter which is a great source of the latest information in this field. There are also freely available YouTube videos and podcasts one could make use of. There are many discord channels for every area – algorithms, MLOPs, software engineering, deployments, etc. and you can join the ones related to your area of interest and expertise. This will help you identify new opportunities and stay ahead of the curve.
• Network and collaborate with other professionals: You can join meetups in your city or area to connect with other professionals in this field to know about the developments and research being done elsewhere. There are a lot of ML conferences and hackathons that happen throughout the year which are a great source of learning as well as networking with other professionals. LinkedIn groups and forums, industry events, and community workshops are also great ways to learn from others and stay up to date with the latest trends in the field.

HackerEarth: Your final word to developers in this stream: What do you developers need to know to excel in data analytics or data security and what are your top 3 expert tips?

Ashutosh: To excel in data analytics, developers should have a strong foundation in math and statistics, as well as programming skills. They should be proficient in using tools and technologies for data manipulation, visualization, and analysis, such as SQL, Python, and R. In addition, they should have strong communication and problem-solving skills, as they will often be working with large and complex datasets and will have to clearly present their findings and recommendations to stakeholders.

Here are my top 3 tips for developers interested in pursuing a career in data analytics:

1. Practice, practice, practice: The best way to improve your skills in data analytics is to get hands-on experience working with real data. This can involve working on personal projects, participating in online hackathons or data science competitions, or taking on internships or freelance projects.
2. Stay up to date: The field of data analytics is constantly evolving. Follow the latest technologies and best practices in order to remain competitive in the job market. This can involve reading industry blogs and news, attending conferences and workshops, and taking online courses to learn new skills.
3. Build a strong network: Networking is an important aspect of any career and is especially important in the field of data analytics. Building relationships with other professionals in the field can help you stay connected to the latest trends and opportunities and can also provide valuable mentorship and guidance as you progress in your career.

HackerEarth: Your final word to recruiters hiring for the role: What specialized tools do you think they should be using, what markers of skill should they be looking for, and how can they improve their own understanding of the domain in order to hire better?

Ashutosh: As a recruiter or hiring manager for data science roles, it can be helpful to use specialized tools and platforms to identify and evaluate candidates. Some options may include:

• Online coding platforms: These allow candidates to complete coding challenges or take technical assessments to demonstrate their skills. Examples include HackerEarth, CodeSignal, and TopCoder.
• Data science-specific job boards: There are several job boards specifically focused on data science roles, such as Kaggle Jobs and Data Science Central. These can be good places to find candidates with relevant experience and skills.
• Resume screening software: Tools like Lever and Jobvite can help automate the resume review process by identifying keywords and qualifications relevant to the role.

Also, read: 10 Tech Recruiting Strategies To Find The Best Tech Talent

In terms of markers of skills, there are a few key areas to focus on when evaluating candidates for data science roles:

• Technical skills: Look for candidates with strong programming skills, as well as experience with data manipulation, visualization, and analysis tools such as SQL, Excel, and data analysis libraries like Pandas and NumPy. Experience with machine learning libraries like sci-kit-learn, TensorFlow, and Keras can also be valuable.
• Problem-solving skills: Data scientists should be able to identify and define problems, develop hypotheses and models, and evaluate the results of their work. Look for candidates who have a track record of successfully tackling data-driven projects and can demonstrate the results they achieved.
• Communication and collaboration skills: Data scientists should be able to clearly articulate their methods and findings to both technical and non-technical audiences, and work effectively as part of a team. Look for candidates who have strong verbal and written communication skills, as well as the ability to work well with others.
• Domain expertise: It can be helpful to look for candidates who have a strong understanding of the specific domain or industry in which they will be working. This can help ensure that they are able to apply their skills and knowledge in a way that is relevant and impactful.

To improve their own understanding of the domain, recruiters can seek out training and education opportunities, such as online courses or industry conferences. They can also stay up to date on the latest developments and best practices in data science by reading articles and publications in the field.

About Ashutosh Kumar: