Talent's Information
-
Location
Pune, India
-
Rate
$13.0 per Hour
-
Experience
8 Year
-
Languages Known
English,Hindi
Available for
Background Verification
40 Hr/Week
About Shekhar
A Data Scientist with 8+ years of experience in the fields of Data Science, Machine Learning, and Data Analysis with a strong foundation in Mathematics, Statistics, and Machine Learning algorithms. I am experienced at exploratory data analysis (EDA), data visualization, and building machine learning models to find insight and action- oriented solutions. Seeking a position as a Data Scientist to solve real-world business problems and advance the organization using data science skills.
Work Experience
Data Scientist
- January 2016 - December 2022 - 7 Year
- India
Projects
Data Validation Automation Using Azure Databrick
- June 2021 - November 2022 - 18 Months
Technologies
Role & Responsibility
-
Dropped the repetitive, time-consuming, tedious process of data analysis & data validation diversified incident data by implementing data validation automation. Reporting the data discrepancies to respective data owners and getting the corrected data leads to delays in the data processing and reporting process.
Developed Databrick Job by using PySpark Notebook which read and validate the data for these markets in a single click and send the automated email alerts to the respective vendors regarding data discrepancies.
All manual interventions and 2.5 hours of manual efforts were substantially reduced, resulting in a streamlined data flow process. enhanced data quality of up to 95%.
ETL Process Automation
- June 2021 - November 2022 - 18 Months
Role & Responsibility
-
Developed an automated process to extract the data from different sources and carry out the data cleaning and wrangling. combining all the data from different markets into a single effective table. loading the consolidated table into the database incrementally.
The above ETL process for incident data was automated using Azure Databricks and Azure Data Factory pipelines, which reduced 90% of manual efforts with enhanced data quality.
Product Quality Prediction In Manufacturing
- October 2019 - May 2021 - 20 Months
Technologies
Role & Responsibility
-
Project is about predicting quality of the part manufactured is ok or not ok. Executed EDA on dataset, feature engineering, feature scaling.
Built ML model & Trained model evaluated based on accuracy, precision, f1 score, Model selection for accuracy enhancement using grid search cv. Selected Random Forest Model for predicting the quality of the product with 90% of Accuracy.
5. Predict Heat Transfer Rate In Heat Exchangers (
- October 2019 - May 2021 - 20 Months
Technologies
Role & Responsibility
-
Project is to build the Regression predictive model to predict the heat transfer rate of the heat exchangers based on different attributes so that it is easy for selection of heat exchanger for specific application
Carry out data pre-processing, feature importance, multi correlations analysis, feature scaling. Built Linear Regression model to predict the Heat Transfer Rate.
6. Customer Booking Cancellation Prediction
- October 2019 - May 2021 - 20 Months
Technologies
Role & Responsibility
-
Project is about classify the customer behavior of cancelling the booking based on number of attributes. Executed Data preprocessing, visualize the data dropped inconsistencies and retained important features.
Created classification models Logistic Regression, Random Forest, Decision Tree, SVM, KNN. Selected ensemble model Random Forest with best accuracy prediction.
NMIET
- July 2017 - October 2019 - 28 Months
Technologies
Role & Responsibility
-
Perform extensive exploratory data analysis (EDA) of all aspect of institute level data like student data, instructor data, course data, and infrastructural data.
Develop dashboards to report the students' enrollment trend, course completion rate, instructors' performance, course curriculum completion, and students' performance in tests.
Develop KPIs to improve students' performance, increase students' enrollment, optimum infrastructure utilization and assignment, and effective timetable preparation.
Build statistical model to predict the Predict the students Enrollment trends at institutes and courses are determined by various aspects of students' data, such as grades, native, family income, ethnicity, and gender.
Conduct the course-wise survey to pinpoint the areas of improvement.
Collect feedback from all stakeholders on a regular basis to ensure continuous improvement.
Severity Prediction For Incident Management Ticke
- June 2021 - December 2022 - 19 Months
Technologies
Role & Responsibility
-
The project is all about building predictive machine learning models to predict the severity of the incidents happening at the shell trading sites, which helps to capture and solve the problem within a minimum amount of time, resulting in reduced downtime and business loss. Performed Data preprocessing, Data Collection, Data Cleaning, Data Wrangling, Feature Engineering, Feature Scaling.
Built different classification Machine Learning Models Logistic Regression, Decision Tree, Random Forest, KNN, SVM. Model evaluation and Model selection based on accuracy, precision, recall, f1 score.
Enhance the model performance by selecting best hyperparameters.
Selected Ensemble learning model to predict the severity of the incident with 92% accuracy.
Soft Skills
Industry Expertise
Education
Mechanical in BE
Pune University- June 2004 - June 2007
