Sr Data Scientist Machine Leaning (ML) Engineer

Sushrutha (Sush) Reddy 410-***-**** adzs5u@r.postjobfree.com LinkedIn

SUMMARY:

Sr Data Scientist with a strong track record of solving challenging business problems through the application of predictive modeling, data processing, data mining algorithms, and data cleaning techniques. I’m adept at creating data visualizations that reveal data-driven insights, providing valuable business intelligence solutions to enhance efficiency for various organizations. My expertise lies in leveraging machine learning (ML) algorithms to develop and train predictive models, conduct feature engineering, optimize hyperparameters, and assess model performance using diverse metrics.

As an active participant in the Python open-source community, I’m deeply passionate about the field of deep reinforcement learning. I consistently stay up to date with the latest developments in machine learning (ML) and Python, ensuring that I bring cutting-edge techniques to my projects. My goal is to deliver high-quality, data-driven solutions to businesses, contributing to their growth and success.

TECHNICAL SKILLS:

Programming Languages: Python, R Studio, C, MYSQL, PySpark, HTML, Oracle DB, PowerShell, Unix & Shell Scripting

Application Tools: Jupyter, Spyder, ML, Statistical tools (Time series, Regression models, PCA etc.), Deep learning, NLP, ServiceNow, AWS, Azure, Excel, Jira, GitHub.

BI Tools: Tableau, Power BI

Platforms/Frameworks: SAP DI (Data Intelligence), Dataiku, Flask, Sage maker, PyTorch, TensorFlow, Kubernetes & HANA

MS Office: MS Word, MS Excel, MS PowerPoint & MS Visio

WORK EXPERIENCE:

Client: Oracle Cerner Corporation

Location: Baltimore, MD (Remote)

Role: Sr Data Scientist/ML Engineer

Timelines: Oct 2022– Present

Roles and Responsibilities:

●Worked in collaboration with the business stakeholders to comprehend the project's requirements and gain insightful information.

●Worked in collaboration with team leads and data engineers to develop project solutions.

●Ensure data quality and integrity while leading data mining and collection procedures.

●Lead machine learning model development projects to forecast business predictions like market trends, customer behavior, and investment opportunities.

●Used Automated Machine Learning (AutoML) within Dataiku to automate the process of model selection and built models.

●Processed large and complex data sets received from different sources.

●Maintained and developed pipelines workflow using recipes of visual, code, plugin for end-to-end data driven projects.

●Developed a CI/CD pipeline to automate the deployment of ML models and applications into Prod and other lower environment.

●Containerize ML applications and services using Docket to ensure the consistency and portability across different environments.

●Used shell scripting for automating various tasks related to deployment, monitoring, and maintenance of ML tasks & systems.

●Used Kubernetes for orchestrating the containers by creating configuration files and worked with creating pods, services, and scaling ML systems.

●Developed automated testing for ML models and pipelines, including data validations and model performances and regression testing.

●Identify areas for business improvement and how to use data modeling to support strategies.

●Communicate findings to enterprise stakeholders and management.

●Created interactive and highly informative Power BI reports using python scripting.

●Used Power BI to collaborate with different users and stakeholders to work on live data sources.

Client: Signify Innovations

Location: Bangalore, India

Role: Data Scientist/Engineer

Timelines: Apr 2021 – Oct 2022

Roles and Responsibilities:

●Developed and prototyped robust and scalable analytical solution models for key business problems using Python and PySpark based on machine learning, data mining, and statistical modeling.

●Created tools and support structures in Python using pandas and NumPy packages for big data collection, data analysis, feature selection, and feature engineering to display and aid in the construction of data engineering pipelines.

●Worked with development teams and business groups to ensure that models could be implemented as part of a delivered solution and replicated across multiple markets/countries.

●Used various "Time Series Forecasting" techniques to predict a surge in product sales, improved accuracy with statistical models (ARIMAX, ARIMA, Regression, ETS), and deployed the project end-to-end on the SAP DI platform to reduce run time from 30 hours to 12 hours.

●Stored model results on HANA tables and directed them to display on SAC dashboards.

●Created various models using ML techniques and statistical algorithms (Gradient Boosting Machine, Sarimax, Elastic-Net, Regression) that resulted in 86% sales based on previous data.

●Used Scikit-learn preprocessing to perform feature engineering tasks such as feature generation, PCA, feature normalization, and label encoding.

●Experimented with and built predictive models to predict sales amount using ensemble methods such as Gradient Boosting trees and neural network by Keras.

●Applied various machine learning algorithms and statistical modeling techniques such as decision trees, text analysis, natural language processing (NLP), supervised and unsupervised regression models, social network analysis, neural networks, deep learning, SVM, and clustering to identify volume using Python's Scikit- learn package.

●Working knowledge of TensorFlow, PyTorch, Keras, and other deep learning frameworks.

●Created an Artificial Neural Network in Python using Tensor Flow to determine the likelihood of customers canceling connections (churn rate prediction).

●Using Sentiment Analysis and Text Analysis, we classified comments from various social networking sites into positive and negative clusters.

●Created map reduce/spark Python modules for Hadoop on AWS for machine learning and predictive analytics.

●Created NLP models for topic extraction, sentiment analysis, and NLP data processing and pattern discovery using the NLTK library.

●"UCM models" were developed in the Market Mix Model (MMM) to ensure the effectiveness of marketing expenditures by gaining access to the historical impacts of marketing campaigns on revenue (Sales).

●Analyzed and improved models for various offline and online marketing channels to achieve a precise and equitable distribution of spending, resulting in 42% of revenue.

●Applied "Random Forest" and "Lasso regression" on 10,000+ features to optimize 3000 features.

●Be accountable for Data Science deliverables.

●Understanding of Agile project management principles.

●Created project requirements and deliverable timelines, which were efficiently executed to meet the deadlines.

Client: PMJ Global Solutions

Location: Bangalore, India

Role: Data Engineer/Scientist

Timelines: Aug 2019 – Apr 2021

Roles and Responsibilities:

●Collecting data from various databases, filtering and cleaning it to gain better insights.

●Worked on data analysis, specifically identifying dataset sources, data sources, metadata data definitions, and data formats. Programming well-designed, testable, efficient code using python language.

●Pandas, NumPy, Sklearn, stats models, seaborn are few of the packages used in model building.

●Worked on data cleaning and ensured data quality consistency and integrity using pandas and NumPy.

●Preliminary data analysis was performed using descriptive statistics, and anomalies such as duplicate removal and missing value imputation were handled.

●Worked with large datasets to break them down (acquired data and cleaned data) Trends were examined using Python visualizations.

●Wrote and optimized SQL queries to perform data extraction to fit the analytical requirements.

●Implemented demand forecasting models, which increased data forecasting and accuracy by 45%.

●Used web scraping techniques to assist the firm in downloading public data such as other companies' descriptions and quotes, resulting in a 50% increase in research efficiency.

●Reduce data latency by standardizing data feeds across systems and streamlining data integration procedures.

●Used Matplotlib to design data visualizations that represent data in a human-readable format.

●Used accuracy measurements such as R-squared, RMSE/MSE to evaluate the performance of various models.

●Used Excel to create various reports using charts and graphs.

●Developed a critical matrix in collaboration with other teams in order to achieve business objectives.

●Created flowcharts, layouts, and documentation to determine requirements and solutions.

●Assisting with deployments as needed.

●Troubleshooting and resolving issues in the production environment.

Client: INNODATATICS

Location: Hyderabad, India

Role: Data Science

Timelines: Sep 2017 – Aug 2019

Roles and Responsibilities:

●Used probability and statistics to gain insights from the data provided.

●Worked on data collection, cleaning, and preparation, as well as exploratory data analysis.

●Acquired knowledge of Supervised and Unsupervised model techniques.

●Hands-on projects using Python and R programming on various regression and machine learning techniques.

●Built a solution to identify the associated products and the ranking of associations using Market Basket analysis.

●Created PowerBI dashboards displaying the top three product associations.

●Learned about various components of the Tableau tool.

●Built a model to predict whether a credit card transaction is fraudulent or not, based on the transaction amount, location and other transaction related data.

●Ensemble, KNN, SVM algorithms are applied to the data to distinguish the new fraud transaction.

●Created different bar graphs, line graphs, heat graphs using the data and captured trends.

●Created a data-driven story and presented the growth to the business.

ACHIEVEMENTS:

●Awarded the "Warrior" title by Signify Innovations India Ltd. in 2021 for actively contributing to and maintaining the team's quality improvement.

●Honored on the “Light Up” recognition platform in 2021 for upholding consistency in work at Signify.

●Led a team of 5 on the Market Basket analysis project, and with proper time management delivered the project on time.

●HEAD and FOUNDER of EVENODDS, the Official College Dance crew of GNITS, founded in 2017.

EDUCATION:

●Bachelor of Technology from JNTU, Hyderabad, India

Contact this candidate