Data-driven problem solver with a passion for Data Science! 🚀 A Master of Data Science graduate from UBC, I bring years of hands-on experience as a Junior Data Scientist, driving impactful insights with clustering, predictive modeling and statistical analysis. Proficient in Python, R, SQL, and a range of tools like scikit-learn, Tableau, and AWS. Eager to leverage data to innovate and make a difference!
SKILLS
Programming: Python(NumPy, Pandas, ScikitLearn, altair, TensorFlow, Selenium), R Tools, Frameworks: OOP, API’s, Django, Docker, AWS, Git Data Visualization: Tableau, Power BI, Plotly Dash Data Analysis: Statistical Analysis, Hypothesis Testing, A/B testing, Web-Scraping, ETL ML: Classification, Regression, Boosting, PCA, Clustering Database: MySQL, NoSQL, MongoDB
EDUCATION
The University of British Columbia
British Columbia, CA
Master of Data Science
Sep 2022 – Jun 2023
- Relevant Coursework: Modelling & Simulation, Data Collection, Data Wrangling, Resampling & Regularization, Predictive Modelling, Bayesian Inference, Supervised, Unsupervised & Semi-Supervised Learning
- Granted exclusive scholarship for outstanding academic and leadership achievements.
American International University-Bangladesh
Dhaka, BD
BSc. in Computer Science & Engineering (CGPA: 3.91/4)
Jan 2018 – Sep 2021
- Relevant Coursework: Artificial Intelligence, Data Warehousing & Data Mining, Computer Vision & Pattern Recognition
WORK EXPERIENCE
Statistics Canada
Ottawa, CA
Data Scientist - Internship
Apr 2023 – Jul 2023
- Utilized clustering algorithms (density-based, distribution-based, centroid-based) to segment proximity measures for various amenities, enhancing the interpretability and practicality of Statistics Canada’s Proximity Measure Database for urban planning and policy-making.
- Used R to perform comprehensive data profiling activities, including data exploration, outlier detection, and statistical analysis, gaining valuable insights into data quality challenges.
- Conducted cluster profiling to extract valuable insights, empowering policymakers and urban planners with data-driven guidance for effective community development initiatives.
PROJECT EXPERIENCE
- Implemented Gaussian Mixture Model (GMM) to cluster customers, identifying high-value segments like ‘High Potential’ customers, resulting in improved marketing efficiency and higher conversion rates.
- Developed a discretization method for numerical variables, categorizing consumers into ‘Low Consumer,’ ‘Frequent Consumer,’ and ‘Biggest Consumer,’ enabling tailored marketing strategies.
- Applied the Apriori algorithm to uncover associations such as frequent fruit consumers being more likely to purchase sweets, optimizing cross-selling strategies and driving revenue growth.
- Launched a user-friendly application on AWS that allows users to input feature variables value for house price prediction, expanding accessibility to predicted house prices.
- Enhanced data reliability through the implementation of data cleaning, standardization, and transformation techniques within machine learning pipelines using scikit-learn to develop accurate models.
- Assessed machine learning models based on RMSE to identify the most optimal predictive performance, ensuring the deployment of the most effective solution. Established a streamlined CI/CD pipeline using GitHub Actions, ensuring efficient AWS deployment and continuous integration of the application.
- Developed a streamlined process for scraping LinkedIn job postings via the LinkedIn API and efficiently refined results through user-specified filters, enhancing the job search experience.
- Integrated the OpenAI API to provide tailored skill and tool suggestions for individual jobs, a strategic approach grounded in job descriptions.
- Created interactive visualizations to showcase the frequency of specific skills and programming languages required in job postings, enabling job seekers to prioritize their skill development effectively.
- Implemented a CI pipeline using GitHub Actions to ensure continuous integration and streamline development workflows.
- Built a machine learning model after reviewing peer reviewed research papers, using SVM and CNN classification that predicts ASD traits with 94% accuracy. Created a dashboard using Tableau for analyzing ASD traits among test takers.
- Achieved recognition by publishing research findings in the esteemed International Journal of Information Technology and Computer Science (IJITCS), contributing to the advancement of knowledge in the field.