Course Work - Machine Learning, Opearting Systems, Parallel Computations, Digital Image Processing, Software Design, Artificial Intelligence, Big Data Analytics, Information and Security, Game Art And Animation
Sreenidhi Institute of Science and Technology
B.S. Computer Science, GPA 3.9
Professional Experiences
Software Developer Intern
Imbuedesk ENS Pvt. Ltd.
April 2018 - August 2018
Worked on a project by contributing mostly to full stack development using Agile methodologies and participated in the whole software development process from design, implementation and delivery
Performed Client side validations and developed web interface and different GUI screens to design UI.
Implemented and tested enhancement feature requests to improve product functionality and extend performance.
Wrote 60% of code, reviewed the other 40%.
Recommended and Implemented new features.
Using Selenium I built a unit testing infrastructure for a client application that reduced the number of bugs reported by the client by 11%.
Implemented various validations on front end and back end as per the requirements.
Wrote e cient SQL Queries to enhance the performance of existing report on Oracle databases. This has improved performance by 50%. Analysed and Identified software defects.
Research and Projects
Hook Underwater – Created a Game [Game Development from scratch | Photoshop | Unity | C#]
Developed a game called Hook Underwater. Developed it by creating all the artwork in photoshop and the coding and development part in Unity using c#. Created and developed everything from scratch and made a complete fun, playable and challenging game.
Full Stack developement project for managing employees data - [ React | HTML | CSS | JS | Springboot | NodeJS | Java | XML | Eclipse | Hibernate | JPA | JSP | JDBC ]
Developed a full stack end-to-end web application which can (Create, Read, Update, Delete) employee system using React on front end and spring boot restful API on back end.
Implemented Micro Services based using Spring Boot.
Implemented Spring rest controllers.
Implemented Spring data JPA for Service layer and used based repository interfaces (JPA/CRUD).
Analysis and extraction of required data from Huge Datasets [ Big Data Analytics | Hadoop | MapReduce | Spark | Cassandra | python]
Performed data analysis by extracting required data from a huge dataset efficiently in very less time using my own logic.
Developed a pyspark code that parallelizes the sequence alignment operation to find the protein sequences which are matching to a single protein sequence by comparing the protein sequence to all the sequences in the database and then performed different measurements by changing different parameters to note down the observations that I observed by taking execution time into consideration.
Found out the hourly concentration of o3 from pollutants of 2013 in Texas dataset by taking some constraints into consideration and found the execution time for different number of executors and observed the results.
Converted the csv data file into parquet and json file formats by developing a pyspark code and compared the file size of the two new formats with the original csv file and compared the execution times as well for hourly concentration using all the three file formats and different number of executors.
Perfomed the above action by importing data into Cassandra as well and observed the difference in execution times to find hourly concentration of o3 using different executors by taking csv file as input vs. by taking Cassandra table as input.
Comparing the performance of artificial neural networks (ANN) with feature selection versus support vector machine (SVM) with feature selection [ Artificial Intelligence | Neural Networks]
Presented different machine learning models using different machine learning algorithms and different feature selection methods to find a best model.
The analysis of the result shows that the model built using ANN and wrapper feature selection outperformed all other models in classifying network traffic correctly with detection rate of 99.02%.
In the ANN model, we experimented with different number of hidden layer and found that the detection success rate varies with the number of hidden layers. After several trial and error methods, we
found best detection rate with 3 hidden layers and 0.1 learning rate. In the wrapper feature selection method, we also used SVM algorithm as classifier.
I believe that these findings and model will certainly contribute to research further in various domainslike for building a detection system that can detect known attacks as well as novel network attacks, for gearbox fault
diagnosis, for coal boiler plants using real time plant data and in various fields of medical diagnosis.
Personal Portfolio of mine [Front End | HTML | CSS]
This is the updated in-detail resume portfolio of mine which adjusts to any device perfectly and I did this using HTML and CSS.
Predicted and generated fuel Quotes for each of the logged-in client based on various parameters using Javascript on front end and PHP on backend and mySQL using Agile methodology with three modules namely Registration Module, Pricing Module and Fuel Quote Generation Module and implemented various validations and authorization features.
Improving the performance of time series data analysis. [Machine Learning, Natural Language Processing, Artificial Intelligence]
Here we proposed a possible solution for the betterment of Time-series data analysis which deals with improving the performance of learning model by the combination of Bi-directional RNN + LSTM = Bi-LSTM + hyperparameter optimization (SMBO i.e., Bayesian optimization with TPE). This combination is proved to show better results when compared to other statistical models like ARIMA and GARCH which could not solve the problem of short term memory.
CUDA Renderer - Parallel Computation Lab using C programming [Parallel Computation, CUDA, C, C++]
Implemented a parallel renderer in CUDA that draws colored circles. In this project lab, I designed and implemented a data structure that can be efficiently constructed and manipulated in parallel. Solved the atomicity and ordering of the circle issues faced in the serial code.
Matrix Multiplication Optimization Using Parallel Computation [C programming]
Optimized the matrix multiplication in Lonestar5 supercomputer using Loop unrolling, vectorization, loop interchange and blocking techniques and increased the average performance percentage to 24.3%.
E-Commerce big data analytics. [Hadoop, MapReduce, Java, HiveQL, Tableau]
Analyzed e-commerce shopping data using hadoop and map reduce OOPS concepts. Used hiveQL in analyzing the data and writing efficient hive SQL queries and also did the same using MapReduce java programs and represented the graph results in Tableau
Image Editor . [Java Programming using OOP Concepts]
Designed and created an image editor desktop application using image processing techniques and java programming using OOPS concepts with many editing options.
Virtual Classroom . [Web Development]
Developed a web based Graphical User Interface using HTML, javascript, CSS, Tomcat, JDBC, Servlets, JSP and mySQL which allows instructors to deliver web based training to geographically dispensed students, employees, business patners and customers.
Personal portfolio . [Responsive Web Designing]
Researched on responsive web designing and various techniques of bootstrapping, building a responsive grid view (which can adjust to any device), jQuery, javascript and applied them in making my project where I designed a personal portfolio using all the above elements.
Paper Publications
How machine learning inspire major change in technology in the near future | International Organization of Scientific Research Journal of Engineering (IOSR-JEN) in September 2018.
IBM Certified Data Architect - Big Data [ From IBM ]
React: Working with APTs [ LinkedIn Learning ]
React.js: Building Interfaces [ LinkedIn Learing ]
Node.js: Essential Training [ LinkedIn Learning ]
Data Science on Google Cloud Platform: Predictive Analytics [ LinkedIn Learning ]
Tableau and R for Analytics Projects [ LinkedIn Learning ]
Honors and Awards
Prime Minister's Scholarship Recipient - Indian Central Government
Codera [ Java, Python ] - Ranked third in the codera intra-campus coding competition among 1250 students.
Telangana Eamcet Rank - [ 4228/3.5 lakh students ] - Secured a top rank in EAMCET which is a state wide examination for admissions into various top engineering colleges across the state.
Co-founder and Head @ Rigolade
Organised talks and workshops
Co-ordinated all the students at the fest
Event Manager @ Sreenidhi Institute of Science and Technology
Organizer of conclave forums and talks
Handled Logistics and Pitched Speakers
Volunteer @ NSS [National Service Scheme - by Indian Govt.]
Created jobs for rural women
Co-organised Cyclone-relief fund raiser
Cleanliness drives and taught rural school kids
Class Committee Representative
Lead the class for four years by being the class represetative for our class from 2015-2019 at Sreenidhi Institute of Science and Technology
