DEC 07, 2021
How to bag that ML internship
OCT 17, 2021
Tips that will help you stay ahead in bagging ML internships
Machine learning is not just a buzzword anymore. It is very much a reality in today’s businesses. There is a huge craze among students and professionals to enter this exciting field, and a stepping stone to that would be an internship – that would give a real-world experience on deploying algorithms to solve business problems.
But how does one go about bagging a machine learning internship? With so much demand to enter this field, the competition to bag these coveted internships is cut-throat. How does one stay ahead in the game to land the internship that could be a stepping stone to their ML careers?
We have you covered. Listed down are some ways to get that ML internship you have been eyeing:
While interning will expose you to a variety of skills and methods, even to bag an ML internship, one needs to come with good mathematical and statistical understanding, programming and business problem-solving skills. Even if you have not worked on ML techniques before, you need not worry. Having a strong base in SQL, R, Python and domain knowledge at your fingertips will make your internship experience smooth.
Priyabrata Mishra, who has used ML techniques at one of his internships
and is currently pursuing an integrated MSc in Mathematics and Computing at BIT Mesra, says, “As an ML internship is all about data, the candidate should be able to play with the given dataset and get to the conclusion based on the problem statement. I think the first and foremost thing needed is basic statistics. Along with it, the candidate should also have knowledge of basic linear algebra, optimization techniques, and a few machine learning Algorithms.
“To handle data, knowledge in one of the programming languages is also needed. At the present time, Python and R are the most used languages in the ML domain, but it depends on the company or the team. DBMS concepts and SQL are good to have. It’s not always required, but in a few instances, the intern might have to pull some required data from the company or client database”, Mishra adds.
If you have worked on independent ML projects on Kaggle and other platforms and can showcase them at sites like GitHub, that will surely catch the attention of the recruiter. Successful projects conducted out of work or educational curriculum shows enthusiasm and eagerness to learn. This makes the recruiting company see potential in you as they think you will go that extra mile to learn if you intern at their place.
Networking can be a key to land ML internships. One should look for machine learning seminars, events, and campus programs to connect with leaders already working in this field. While networking with the leaders, try to understand the particular area of machine learning they work on and see if that area excites you and motivates you to work too. Do not blindly apply to ML internships that do not cater to your career path.
Elman Mansimov, an Applied Scientist at Amazon Web Services, resonated with this idea in a recent tweet. He asked aspirants to work on identifying people whose work they admire and want to work with. He said, “You should already have a rough list of people in mind based on your interests. Many folks in ML I cold-emailed were all friendly and got back to me at some point.”
While big names might seem desirable for many, sometimes, startups can provide a steeper learning curve. While it is perfectly fine to intern at either of the two if you get to work in an area of your choice, bigger names might require advanced degrees to just apply. Startups might have a slightly lower entry barrier in terms of educational qualifications but can accelerate the learning process.
Dipyaman Sanyal, who runs Dono Consulting, a boutique quantitative analytics and financial modelling firm, says, “I might be biased since I run a small firm, but I’d strongly recommend interning at a small firm with the right kind of focus. Our interns work on projects that 3rd-4th year analysts in bigger firms won’t have access to. It also gives you a complete end-to-end picture and access to leaders who have seen quite a bit. However, you must have the mindset to learn and juggle to be a part of a smaller firm. It’s not easy if you have a mindset of just ‘doing your job’.”
Often, we try to overpitch ourselves while applying for an internship or a job. If you have not worked on a particular ML technique or you do not have knowledge about a particular statistical concept, be honest about it. The hiring managers at the companies and startups can easily detect whether you are actually skilled at a particular technique you are boasting about or not.
Once you get that ML internship, the goal should be to convert that into a full-time role. Venkat Raman of Aryma Labs says that one can increase the odds of converting the internship by demonstrating that one is a quick learner and by being inquisitive. He adds, “Make some significant contribution in the project that results in some tangible results for the company.”
With competition in the ML space growing every day, along with the listed tips, the intense desire to learn each day is a must to survive in this field. Data science, ML, AI and other emerging technologies are evolving every day. One has to keep themselves constantly updated with these changes to stay in this field for the long haul.