My resume
SAI VIKHYAT PAREPALLI
------------ | ------------ |Charlotte, North Carolina-28262 | ------------/sparepal | ------------/in/sparepal/
SKILLS
Programming languages: Python, R, Java, PL/SQL, C and C++.
Libraries- Scikit-learn, Tensor Flow, Keras, Numpy, Matplotlib, SciPy and Pandas.
Tools/Computing platforms/Version control: Anaconda, AWS, Hadoop, Map-reduce, R-studio, Jupyter notebook, D3, JavaScript, Spyder, GitHub
Soft skills-Collaboration, Teamwork, communication, critical thinking, independent, strong desire to learn and grow.
EDUCATION
Master of Science at University of North Carolina at Charlotte (UNCC) (January 2017 – December 2018) 3.6/4.0
Computer Science with Data Science concentration, Graduation date- December 2018
• Relevant coursework -, Database Systems, Machine Learning, Knowledge discovery in databases, Cloud computing for data analysis, Deep Learning with Tensor Flow, Computer Vision, Algorithms and Data Structures.
Bachelor of Science at Vellore Institute of technology (VIT) (June 2012 – May 2016) 3.6/4.0
Computer Science-Graduation date – May 2016
• Relevant coursework- Database systems, Soft computing, Data warehousing and data mining, Probability and statistics, Linear Algebra, Multi-variable calculus, Algorithm design and analysis, Data structures & algorithms.
PROFESSIONAL EXPERIENCE
Volunteer Research assistant under Dr. Wlodek Zadrozny- Data science & Business analytics (January 2018 – Current)
• Mathematical look and say sequence generation using Machine learning approach.
Graduate level Teaching assistant UNC Charlotte under Dr. Wlodek Zadrozny- Data science & Business analytics (July 2018 – December 2018)
• Negotiated the syllabus to be posted while assisting Professor in delivering concepts regarding Big Data and Modern Data Science systems.
• Reinforcing concepts to students by helping them with the school programming assignments & projects while providing feedback & grades.
Data analyst intern at Inforica Pvt ltd (July 2016 – December 2016)
• Collected and analyzed data from clients, advising them in potential betterment of their usage.
• Translating the numbers and the labels of the data given to have informed better business decisions.
ACADEMIC PROJECTS
American hand gesture recognition using TensorFlow (January 2019 –February 2019)
• Increased training speed of model by using mini-batch gradient descent using random sampling.
• Achieved a 94.2% accuracy for the multi class classification using CNN’s compared to 84%-88% achievers on Kaggle kernels.
• Developed using TensorFlow framework additionally including frozen graphs for deployment on Android using Tensorflow SDK & NDK.
Predicting chance of admit for graduate admission (September 2018 –December 2018)
• Predicting the chance of admit from various features of students such as GPA, TOEFL and GRE scores, LOR, Research and SOP.
• Found out that the CGPA was the highest correlating variable.
• Ran the dataset though Linear regression and Random Forest and then using cross-validation and tuned using GridSearchCV.
Text document comparison - language processing (July 2018 –September 2018)
• Word2Vec implementation comparing wiki documents using cosine similarity and TSNE. Pre-process and clean text; extracting tokens.
• Programming language: Python, IDE-Spyder. Libraries used- Numpy, Pandas, TensorFlow, Genism and MatPlotLib.
Time series forecasting of solar energy generated using sensory data (August 2018 – October 2018)
• Improved RMSE by 40% using LSTM instead of ANN and traditional ARIMA models after applying feature selection.
• Helped better processing of data by model by experimenting with the sliding window size and various hyper-parameters.
• Implemented using the Keras Library. IDE-Spyder & Jupyter notebook. Libraries- Keras, Numpy, Pandas, GridSearchCV and MatPlotLib.
Machine Learning algorithm performance comparison on various datasets from UCI ML repository (March 2018 –June 2018)
• Exercising with different algorithms on Binary classification of Tumors, Breast Cancer and Wine quality regression.
• Observed ANN, SVM, Random forest and Ensemble learning produces equally best results while comparing the running times of each.
• Ran PCA on the datasets to observe the linear separability and visualization on 2D. Language: Python. IDE- Jupyter notebook & Spyder.
Gift Registry High level database design (January 2018 –March 2018)
• Handled high volume of data using proper normalization, planning, design and naming conventions,
• Optimized the implementation of the database design of the Registry application by implementing indexes, views, stored procedures, functions and triggers on a 4NF high level database design. Implemented using phpMyAdmin and MySQL server.
Machine Learning library (July 2017 –November 2017)
• Scratch and step-by-step implementation of algorithms-PCA, LDA, K-Means, Hierarchical clustering, Linear & Logistic regression & ANN.¬
• Datasets with same algorithms were run by Scikit-learn which produced similar results.
• Built using only the basic Numpy and Pandas and MatPlotLib Libraries. Language: Python. IDE- Jupyter notebook & Spyder.
Coura (Software development project) (February 2017 –May 2017)
• Developed website recommendation feature different for each user based on registration(s) & search history using content-based filtering.
• Technologies used-AngularJS as the front-end framework, database on MySQL server and backend-Java with Spring MVC framework.
CERTIFICATIONS
• Python for Data Science – Coursera, Machine Learning A to Z -Udemy and Deep learning - Udemy
SAI VIKHYAT PAREPALLI
------------ | ------------ |Charlotte, North Carolina-28262 | ------------/sparepal | ------------/in/sparepal/
SKILLS
Programming languages: Python, R, Java, PL/SQL, C and C++.
Libraries- Scikit-learn, Tensor Flow, Keras, Numpy, Matplotlib, SciPy and Pandas.
Tools/Computing platforms/Version control: Anaconda, AWS, Hadoop, Map-reduce, R-studio, Jupyter notebook, D3, JavaScript, Spyder, GitHub
Soft skills-Collaboration, Teamwork, communication, critical thinking, independent, strong desire to learn and grow.
EDUCATION
Master of Science at University of North Carolina at Charlotte (UNCC) (January 2017 – December 2018) 3.6/4.0
Computer Science with Data Science concentration, Graduation date- December 2018
• Relevant coursework -, Database Systems, Machine Learning, Knowledge discovery in databases, Cloud computing for data analysis, Deep Learning with Tensor Flow, Computer Vision, Algorithms and Data Structures.
Bachelor of Science at Vellore Institute of technology (VIT) (June 2012 – May 2016) 3.6/4.0
Computer Science-Graduation date – May 2016
• Relevant coursework- Database systems, Soft computing, Data warehousing and data mining, Probability and statistics, Linear Algebra, Multi-variable calculus, Algorithm design and analysis, Data structures & algorithms.
PROFESSIONAL EXPERIENCE
Volunteer Research assistant under Dr. Wlodek Zadrozny- Data science & Business analytics (January 2018 – Current)
• Mathematical look and say sequence generation using Machine learning approach.
Graduate level Teaching assistant UNC Charlotte under Dr. Wlodek Zadrozny- Data science & Business analytics (July 2018 – December 2018)
• Negotiated the syllabus to be posted while assisting Professor in delivering concepts regarding Big Data and Modern Data Science systems.
• Reinforcing concepts to students by helping them with the school programming assignments & projects while providing feedback & grades.
Data analyst intern at Inforica Pvt ltd (July 2016 – December 2016)
• Collected and analyzed data from clients, advising them in potential betterment of their usage.
• Translating the numbers and the labels of the data given to have informed better business decisions.
ACADEMIC PROJECTS
American hand gesture recognition using TensorFlow (January 2019 –February 2019)
• Increased training speed of model by using mini-batch gradient descent using random sampling.
• Achieved a 94.2% accuracy for the multi class classification using CNN’s compared to 84%-88% achievers on Kaggle kernels.
• Developed using TensorFlow framework additionally including frozen graphs for deployment on Android using Tensorflow SDK & NDK.
Predicting chance of admit for graduate admission (September 2018 –December 2018)
• Predicting the chance of admit from various features of students such as GPA, TOEFL and GRE scores, LOR, Research and SOP.
• Found out that the CGPA was the highest correlating variable.
• Ran the dataset though Linear regression and Random Forest and then using cross-validation and tuned using GridSearchCV.
Text document comparison - language processing (July 2018 –September 2018)
• Word2Vec implementation comparing wiki documents using cosine similarity and TSNE. Pre-process and clean text; extracting tokens.
• Programming language: Python, IDE-Spyder. Libraries used- Numpy, Pandas, TensorFlow, Genism and MatPlotLib.
Time series forecasting of solar energy generated using sensory data (August 2018 – October 2018)
• Improved RMSE by 40% using LSTM instead of ANN and traditional ARIMA models after applying feature selection.
• Helped better processing of data by model by experimenting with the sliding window size and various hyper-parameters.
• Implemented using the Keras Library. IDE-Spyder & Jupyter notebook. Libraries- Keras, Numpy, Pandas, GridSearchCV and MatPlotLib.
Machine Learning algorithm performance comparison on various datasets from UCI ML repository (March 2018 –June 2018)
• Exercising with different algorithms on Binary classification of Tumors, Breast Cancer and Wine quality regression.
• Observed ANN, SVM, Random forest and Ensemble learning produces equally best results while comparing the running times of each.
• Ran PCA on the datasets to observe the linear separability and visualization on 2D. Language: Python. IDE- Jupyter notebook & Spyder.
Gift Registry High level database design (January 2018 –March 2018)
• Handled high volume of data using proper normalization, planning, design and naming conventions,
• Optimized the implementation of the database design of the Registry application by implementing indexes, views, stored procedures, functions and triggers on a 4NF high level database design. Implemented using phpMyAdmin and MySQL server.
Machine Learning library (July 2017 –November 2017)
• Scratch and step-by-step implementation of algorithms-PCA, LDA, K-Means, Hierarchical clustering, Linear & Logistic regression & ANN.¬
• Datasets with same algorithms were run by Scikit-learn which produced similar results.
• Built using only the basic Numpy and Pandas and MatPlotLib Libraries. Language: Python. IDE- Jupyter notebook & Spyder.
Coura (Software development project) (February 2017 –May 2017)
• Developed website recommendation feature different for each user based on registration(s) & search history using content-based filtering.
• Technologies used-AngularJS as the front-end framework, database on MySQL server and backend-Java with Spring MVC framework.
CERTIFICATIONS
• Python for Data Science – Coursera, Machine Learning A to Z -Udemy and Deep learning - Udemy