About

Karan Sood is a Master's student at Lakehead University with a passion for data science. He is an enthusiastic learner and is fascinated by numbers, always striving to find truth in the information chaos. Karan is highly skilled in fields of Data Mining and Machine Learning, which he believes have opened up new possibilities for discovering valuable insights from data. Karan has gained valuable experience in various aspects of data science, including mining, analysis, and modeling, through his work as a research advanced analyst. He has also worked on personal projects in various domains, such as health-care and finance, which has helped him develop an even better understanding of various machine learning and deep learning techniques. Currently, Karan is working as a research advanced analyst at Roche, where he is responsible for developing prototypes to calculate the frequency of specific deviations in the manufacturing process. He has also used Google AutoML to extract named entities from deviation records while handling class imbalance with techniques such as under-sampling, over-sampling, and SMOTE. Karan has also used SentenceTransformer to calculate similarity between the user query and sentences extracted from deviation records to expedite search for deviations in the drug manufacturing process. Moreover, he has discovered important patterns in German text datasets using clustering algorithms such as KNN and DBSCAN, which helped to identify new drugs for certain diseases. Before joining Roche, Karan worked as a Data Analyst/Scientist at Niramal Health Analytica, where he performed exploratory data analysis on breast cancer datasets to understand various parameters that affect the severity of tumors. He also compared various feature selection techniques such as RFE (recursive feature elimination) using the Random Forest classifier and PCA (principal component analysis) to identify the best features. Karan built a Flask app to classify a tumor as benign or malignant based on selected features. Karan has a Bachelor's degree in Technology from GGSIPU and a Master's degree in Computer Science from Lakehead University. He also completed his high school education from Doon Public School. Karan's tech stack includes Research Scientist, Data Scientist, ML, Deep Learning, NLP, Search, Android, and Flask. With 4.53 years of relevant experience, Karan is actively searching for full-time/contract roles in Data Science where he can combine his knowledge of Machine Learning, Deep Learning, and Natural Language Processing and contribute towards creating a better and more sustainable data-driven economy.