About

Nabajit Dey is a Senior Engineer (Data Science) with over 3 years of industry experience, including 2 years of experience in data science using ML algorithms and deep learning methodologies. He is proficient in Python and has worked with libraries such as Sklearn, Tensorflow, and Pandas. Nabajit has successfully designed and presented products on demand forecasting, demand segmentation, and classification for clients in the airline and hospitality domains. Nabajit's technical expertise includes Python (e.g., scikit-learn, pandas, numpy, matplotlib), SQL, Jupyter Notebook, VS Code, Power BI, QuickSight, and AWS. He is skilled in the data science pipeline (data analysis, visualization, statistical modeling, interpretation), time series, ML algorithms, and artificial neural network methodologies. Currently serving his notice period at ibs software, Nabajit is working as a Senior Product Engineer (Data Science) on the Dynamic Revenue Management System. He has set up meetings with senior leaderships of an airline giant in Japan to explain the technical aspects of DRMS and important ML algorithms. He has also analyzed and validated data sent by the client. Nabajit has also worked on a POC for demand forecasting for airlines, where he set up a forecasting problem statement for demands in airline bookings with respect to days before departure. He analyzed the time series and used ARIMA, XGBoost, and deep learning models to forecast the load factors, achieving an MAPE of 7.53. In a previous role as a Software Engineer (Data Science) at ibs software, Nabajit developed a pipeline on AWS EC2 and S3 for hospitality forecasting. He also researched various methodologies for clustering demands of various hotels in clients' platforms and implemented hierarchical clustering on multiple time series with DTW as the distance measure and Fourier Transform for noise and dimensionality reduction. Nabajit has also analyzed pre-COVID and post-COVID demands in a hospitality platform for various high-end hotels in the US, utilizing Python to implement ARIMA, CART, boosting, and CNN algorithms for forecasting daily net reservation counts. A walk-forward model in CNN for forecasting the next 30 days booking counts gave him a MAPE of 8.7. Nabajit holds a Bachelor's degree in Electronics and Telecommunication from Kalinga Institute of Industrial Technology, Bhubaneswar. He also completed his School/High School from Kendriya Vidyalaya IIMC Joka.