Data Scientist
Gandhinagar, Gujarat, India
• Provided data science consulting services to several edtech companies, assisting them in solving complex problems through advanced analytics and machine learning resulting in the unlocking of more than $25 million in value.• Led research (4 conference publications) aimed at helping the company establish solid foundations in the fields of data mining, natural language processing, data visualization, machine learning, and algorithm optimization.• Developed a suite of quantitative and graphical algorithms in R to detect cheating patterns in national-level exams, analyzing exam data of over 5 million students. Detected 5%-7% students across various national-level exams who have very high probability of recourse to unfair means.• Analyzed and optimized existing implementation of response similarity measuring statistical algorithm with an average speedup of 50x.• Implemented Data Matching/Entity Resolution/Record Linkage algorithm and R Shiny application for efficient data integration and mapping across multiple data systems.• Provided insights on server load trends, created a predictive model for providing prediction on maximum concurrent users, resolving frequent crashes, and optimizing cloud operations cost.