Well-qualified Data Scientist experienced working with vast data sets to break down information, gather relevant points and solve advanced business problems. Skilled in predictive modeling, data mining and hypothetical testing. Offering 4 years of experience in improving business operations.
Overview
4
4
years of professional experience
6
6
Certification
Work History
Senior Data Scientist
Ernst & Young Global Limited
Bengaluru, India
12.2021 - 07.2022
Designed predictive modeling framework for world's 3rd largest FMCG, which achieved an accuracy of over 90% and bias under 15%, outperforming previous demand planning techniques across all categories in the Nordics market.
Delivered Gen-2 version (70% Volume and SKU coverage) for the Norwegian market within 15 days, benchmarking among European, South-East Asian, and South American markets.
Built automated data discovery streams and data pipelines and implemented code modularization along with E2E testing for production on Azure Data-Lake utilizing Apache Spark and Python.
Leveraged highly intermittent time-series data sources to extract deep insights on demand, such as previous sales, interior trends, external trends, events, and promotions.
Performed regression and variation analysis across different FMCG categories and conducted exploratory data analysis on variables, such as product volume and market cannibalization.
Machine Learning Engineer
Applied Roots
Hyderabad, India
09.2019 - 11.2021
Designed, released, and maintained six python based Machine Learning and Deep Learning projects in Applied Roots (an e-learning platform), along with mentoring and evaluating assignments of over 8000 students.
Contributed to 100% year-over-year revenue growth by creating operational strategies and technical development of the platform.
Designed plagiarism checker for student coursework by leveraging various NLP and Deep Learning techniques such as Word-to-Vector, Code-to-Vector, Pre-trained Language Models, and LSTM.
Research Project Assistant
Stellenbosch University
Stellenbosch, South Africa
03.2021 - 11.2021
Designed, trained, fine-tuned, and validated several deep-learning and transfer-learning based models such as MLP, RNN, CNN, VGG-16, and RESNET-50 for predicting COVID-19 using patient's cough audio.
Achieved specificity, sensitivity, AUC, and accuracy scores of 0.9, 0.93, 0.96, and 92.78% resp. Using RESNET-50 model, making COVID-19 diagnostic through this tool on par with the RT-PCR test.
Developed web-application using Python and Javascript for users to get diagnosed with COVID-19 within 5 seconds through their cough audio.
Freelance Data Scientist
ViVi Green Innovators (Viscon Group)
Amsterdam, MO
10.2020 - 01.2021
Utilized python and image processing for objectskeletonization, localization, and unsupervised machine learning (DBSCAN clustering) to develop and test a machine vision and robotic process automation-based product for extraction of plant cuttings, used in horticulture for vegetative (asexual) propagation.
Developed and delivered a prototype within 70% of deadline duration with 5/5 client satisfaction.
Education
Master of Science - Business Analytics (Data Science Specialization)
University of Texas At Dallas
Dallas, TX
05.2024
Bachelor of Technology - Electronics And Communications Engineering
JSS Academy of Technical Education
Noida, India
06.2018
Skills
Data Analysis & Data Mining: Scikit-Learn
Probability & Statistics, Multivariate Calculus & Linear Algebra
Machine Learning & Deep Learning
Tensorflow, Keras, PyTorch
Python, MySQL, MongoDB, PySpark
Computer Vision, Natural Language Processing (NLP)