Tushar Deep

Data Scientist
[Link]@[Link] +91-6268065740 LinkedIN Github Bangalore, India
Data Science enthusiast with a deep understanding of the latest ML and DL techniques and technologies. Proven ability to solve complex
problems using ML & DL, and to deliver results that meet or exceed business expectations. Excited to join a team that is pushing the
boundaries of data science and artificial intelligence.

SKILLS
 Machine Learning  Deep Learning- ANN,CNN,RNN,OpenCV,BERT
 NLP  Deployment (AWS, GCP, Azure)
 Python Programming  Statisics
 Time Series  SQL
 Numpy, Pandas, Matplotlib  PyTorch/ TensorFlow Frameworks
 Excel

WORK EXPERIENCE
Cognizant Technology Solution Pvt. Ltd., Bengaluru 07-2021 to Present
Technical Lead
 Contributing to the development and maintenance of machine learning models. Played a crucial role in debugging code formachine learning models,
ensuring their proper functionality and accuracy.

 Conducted statistical data analysis on large datasets, extracting meaningful insights and identifying trends to inform decision-making processes.

 Assisting in feature engineering, data preprocessing, and model validation to optimize model performance and enhance predictive capabilities.

 Collaborated with a team of data scientists and engineers to build predictive machine learning models, leveraging algorithms such as linear
regression,decision trees, or SVM etc.

 Actively participated in team discussions, sharing ideas, andcontributing to the improvement of the overall ML workflflow and project outcomes.

PERSONAL PROJECTS

Gem Price Prediction Application

 Developed a highly accurate ML algorithm using Linear Regression to predict the gem price of a given gem. The dataset has features which
decribe the physical characteristics of gems.
 Implemented a scalable and automated pipeline for data ingestion, feature engineering, and model training using Python and popular libraries like
scikit-learn, pandas, matplotlib.

 Employed best practices in data handling, model evaluation, and performance optimization to ensure accurate predictions and efficient
processing.

 Leveraged AWS services such as EC2, S3, and Lambda to deploy the machine learning application in a cloud environment, ensuring high availability
and scalability.

 Integrated the application with a user-friendly interface to allow users to interact with the model and obtain predictions in real-time.

 Developed a user-friendly web interface to allow users to input relevant customer information and obtain predictions on credit card payment outcomes in
real-time.
Credit Card Payment Prediction Application
 Developed an end-to-end machine learning application for credit card payment prediction, enabling accurate forecasting of customer payment behavior.
 Implemented the application using Python, leveraging popular libraries such as scikit-learn, pandas, data exploration and preprocessing to handle missing
values, outliers, and feature engineering to extract meaningful insights from the dataset. and Flask for model training, data preprocessing, and API
development.
 Employed various machine learning algorithms, including logistic regression, decision trees, and random forests, to build predictive models and used CV
for tuning the hyperparameter that forecasted credit card payment outcomes.
 Developed a user-friendly web interface to allow users to input relevant customer information and obtain predictions on credit card payment outcomes in
real-time.

Zomato Order Delivery Time Prediction End to End ML Application

 Built a data ingestion module to collect and preprocess data from various sources, ensuring its readiness for further processing.

 Implemented data transformation steps, including feature engineering, handling missing values, and scaling, to prepare the data for modeling.

 Developed a machine learning model trainer component, utilizing the Random Forest algorithm for delivery time prediction. Conducted hyperparameter
tuning and performance evaluation.

 Created a prediction pipeline to enable real-time predictions on new data, applying the same preprocessing steps as the data transformation module.

 Dockerized the complete application setup, allowing for easy deployment and integration into larger systems.

 Utilized Python, Pandas, Scikit-learn, and Docker as the core technologies for the project.

SMS Spam Classifier NLP Model

 Developed an SMS spam classification model using NLP techniques and Multinomial Naive Bayes algorithm, achieving an impressive
accuracy of 97%.
 Applied text preprocessing techniques such as tokenization, stemming, and stop word removal to clean and prepare the SMS data for
analysis.
 Engineered relevant features from the SMS text, such as word frequency and n-gram representations, to train the Multinomial Naive Bayes
model.
 Leveraged the NLTK library and scikit-learn's implementation of Multinomial Naive Bayes to build and train the classification model.
 Conducted extensive model evaluation and performance tuning, optimizing hyperparameters to achieve the high accuracy of 97% on the
SMS spam classification task.
 Documented and presented the project findings, demonstrating the effectiveness of the NLP-based approach for SMS spam detection and
its potential applications in real-world scenarios.

INTERNSHIPS

Data Analysis and Reporting

JSW Steel Pvt. Ltd, Vijayanagara
 Completed an internship at JSW, a leading steel manufacturing company, where I gained hands-on experience in data analysis and reporting.

 Utilized various tools and techniques, including Excel, SQL, and data visualization software, to analyze and visualize complex data sets.

 Conducted statistical analysis to identify trends, patterns, and correlations, contributing to data-driven decision-making processes.

 Created comprehensive reports and dashboards to communicate key findings and actionable recommendations to stakeholders.

 Collaborated with cross-functional teams to identify key performance indicators (KPIs) and implemented data-driven strategies to optimize operational
efficiency.

 Actively participated in team meetings, sharing insights and contributing to ongoing projects, fostering a collaborative work environment
Virtual Internship - Data Analysis and Visualization
Accenture | Forage

 Completed a virtual internship offered by Accenture through Forage, a leading platform for virtual internships.

 Applied data analysis and visualization techniques during the internship, utilizing Jupyter Notebook for Exploratory Data Analysis (EDA) on provided
datasets.

 Developed proficiency in data manipulation, visualization, and analysis using Jupyter Notebook and relevant tools and technologies.

 Gained comprehensive insights into the role of a Data Analyst in organizational settings and acquired a strong understanding of the required skills and
responsibilities.

 Successfully addressed client-provided questions by leveraging data analysis skills, presenting findings through effective data visualization, and offering
comprehensive answers based on insightful analysis.

 Demonstrated problem-solving abilities by tackling a client-defined problem statement, providing a solution supported by analyzed data, and effectively
communicating the outcomes.

 Overcame challenges encountered throughout the analysis process, displaying adaptability, critical thinking, and the ability to troubleshoot and deliver
accurate and meaningful insights.

EDUCATION
B.E. (Electrical and Electronics Engineering)
Ballari Institute of Technology and Management (VTU, Belgaum)
CGPA = 7.3/10
12th
Jindal Pre-University College (KSEEB)
Percentage = 76.16
10th
Jindal Adarsha Vidyalaya (KSEEB)
Percentage = 82.72

ACHIEVEMENTS
Received recognition in 2022 and 2023 from the Project Director for excellence in debugging code, statistical data analysis, and contributions to building
predictive machine learning models.

CERTIFICATE
 Python for Beginners - Udemy
 Advanced Python for Business Intelligence and Data Analysis - Udemy
 SQL and MySQL for Data Analytics - Udemy
 Google Data Analytics Professional - Coursera

LANGUAGES
 English  Hindi  Kannda/Telugu
Full Professional Profificiency Full Professional Profificiency Full Professional Profificiency

INTERESTS
 Rapping  Cooking  Youtube content creation  Music Composition