About the author: Shantanu Sham Hule is a third-year engineering student at Marathwada Mitra Mandal’s College Of Engineering, Pune. He shares his story of self-learning data science and forging a career path in it.
I first learned about data science and machine learning in my first year of college when I was learning Python. Although it was not a part of my college curriculum, I was curious about it, so I started learning data science concepts from YouTube. Since I was interested in this field, I also started looking at the skill set needed for an internship in data science. When I realised that I did not have skills like deep learning and computer vision, I started working on these before applying for internships.
The first internship that I landed through Internshala was with Nikhil S. In the beginning, I was uncertain whether I would be able to get selected through Internshala. So when I got the internship, I felt more confident. It was a one-month long internship wherein I worked on image segmentation. My next internship was as a machine learning and artificial intelligence intern at Rainet Technology Pvt. Ltd. Although I had been selected for another internship at the same time, I decided to go with this one as they had a data science team, which I thought would be a good learning experience for me.
The selection process consisted of an assignment and an interview round. They had given me data for three skin diseases, and I needed to create a model to classify these three categories. So if an image was given to the model, it should be able to predict which skin disease it was. After clearing this round, I attended a telephonic interview wherein I was asked the following questions:
I. What is a neuron?
II. What are transformers?
III. What are convolutional neural networks?
This was a work-from-home internship, and I was interning from Monday to Friday. During my three-month stint, I was assigned tasks daily for four projects that I got to work on. For example, there was a banking project wherein we had to do feature engineering and create models. There were thirty features and we needed to shortlist only twelve features. This was determined by how much dependency each feature impacted the output. Initially, I found the project difficult, but my mentor helped me understand how I could research better to complete the task. I also worked with other teams on a computer vision project, a research paper related to Covid-19, and a natural language processing (NLP) project.
The knowledge that I gained from this internship has helped me in undertaking my own projects such as the detection of diseases from CT images. I am also taking advanced courses in computer vision to keep upskilling.
Do you want to try your hands at data science, too? Then, check out these data science internships.
If you want to learn all the necessary skills before applying for internships or jobs, then check out Internshala’s Job Oriented Data Science course.