MADHUPRIYA PAL
**********.****@********.*** www.linkedin.com/in/madhupriya6 www.github.com/madhupriya91 469-***-**** EDUCATION
The University of Texas at Dallas, Richardson, Texas May 2018 Master of Science, Computer Science GPA 3.66/4
Kalinga Institute of Industrial Technology May 2013 Bachelor of Technology, Electronics and Telecommunication GPA 8.9/10 TECHNICAL SKILLS
Programming Languages: Java, Python (Pandas, Scikit learn), C, C++, R, MATLAB,PL/SQL Web Based Technologies: HTML, CSS, Javascript, PHP, AJAX Database: Oracle, MySQL
Tools: CVS version control, GitHub, Juniper Regression Manager, OpenCV, Eclipse IDE, Microsoft Excel(v-lookup), Microsoft Power Point, Microsoft Word, JetBrains PyCharm IDE
Operating Systems: Windows, Ubuntu, Solaris, Unix
EXPERIENCE
Aricent, Sr Software Engineer Sep 2013 – Jul 2016
• Trained the team of 60 members the networking protocols and the architecture of newly introduced Juniper switches.
• Analysed and fixed multiple script failure issues caused by the migration from old to the newly introduced highly unstable switches, which resulted in significant reduction in the number of test case failures from 250 to 80 in a regression run of 1000 test cases.
• Scripted test cases in PERL for capturing the bugs reported by customer by recreating the issues manually in the reported release.
• Assigned and allocated the 20 team members the work of taking weekly training sessions. PROJECTS
Machine Learning: Mercari Price Suggestion Kaggle Challenge Dec 2017 – Present
• Implemented various data visualization techniques in Python and performed EDA (Exploratory Data Analysis).
• Pre-processed the data for handling missing values, numeric values and augmented it with engineered features.
• Implementing different regression models like Gradient boosting and Random Forest for price suggestion of various products. Skillset: Python, Pandas, Scikit Learn, Machine Learning models Machine Learning: Dengue Case Prediction Nov 2017 – Dec 2017
• Performed time series forecasting using various regression models like Neural Network, Random forest regressor in Python using Scikit learn packages
• Pre-processed the data and compared the performance in terms of accuracy, median absolute error and mean absolute error. Best results were observed with Gradient Boosting, with accuracy of .8, Mean Absolute Error of 9.6 and Median Absolute Error of 9.3. Skillset: Python, Pandas, Scikit Learn, Machine Learning models Machine Learning: Comparison of Models Nov 2017 – Nov 2017
• Compared the performance of the models: Decision tree, neural networks, SVM, Naïve Bayes and Logistic regression on the Breast Cancer Wisconsin dataset in Python.
• Observed Neural Net to give the best results with accuracy of .98 and F1 score of .97 followed by Logistic regression model and SVM. Skillset: Python, Pandas, Scikit Learn, Machine Learning models Cloud Computing: SaaS Application on a PaaS platform Aug 2017 – Dec 2017
• Designed an application to convert the standalone RoboCode to a web based application, and enhanced it to become a Web based gaming system by deploying it on the open source cloud platform: Cloud Foundry.
• Deployed Oauth2 protocol in Java for authorization and authentication of the Robocode application while trying to access the user’s information from the cloud. Skillset: Cloud Foundry, Java, Unix, Oauth2,Apache OLTU Socket Programming: Chat Application Mar 2017 – Apr 2017
• Developed a server client based chat application in Java using socket programming. Multiple clients can send and receive messages to one or many clients. The server stores all details of each client in a tree map and stores and displays the messages that each user receives.
• Successfully created an effective and user-friendly chat application where several clients were able to connect to the server using their own sockets and fetch or send new messages to other clients.
Skillset: Java Sockets
Parallel Processing: Post Office simulation Feb 2017 – Mar 2017
• Established a multi-threading environment for simulating a post office with 50 customers and three post office workers using minimum mutual exclusion to achieve maximum parallelism (synchronization primitives were coded from scratch).
• Successfully achieved parallelism by executing different post office tasks for 3 customers in parallel at a time by using Runnable interface and synchronization techniques using semaphores and mutex.
Skillset: Multithreading in Java, Runnable interface Application Development: Quick Checkout Mar 2016 – Mar 2016
• Created a java application to avoid wait time at checkout counters.
• It uses phone’s camera to scan the barcodes of the items, submits the details to the store server, generates the bill and updates the inventory database as per the purchase.
Skillset: Java
ACTIVITIES
• Conducted awareness camps for parents in tribal areas in Bhubaneswar (Odisha, India) as a volunteer of National Service Scheme (NSS) India, instilling in them the need of education for their children.
• Member - Women Who Compute (WWC) 2016 - Present
• Member - Association for Computing Machinery at UTD (ACM). 2016- Present