*** ***** ******, *** ** MRUNMAYEE SHINDE 774-***-****

Lowell, MA ac59wb@r.postjobfree.com

OBJECTIVE: Looking for full-time opportunities as Entry Level Data Scientist/ Machine Learning Engineer. EDUCATION:

University of Massachusetts Lowell, Lowell, Massachusetts September 2016- May 2018

M.S in Computer Science, GPA: 3.4

Graduate Coursework: Natural Language Processing, Machine Learning, Operating System, Internet Web Systems, Data Mining, Database, Algorithms, Foundations of Computer Science, Internet of Things. University of Pune, India June 2011- May 2015

B.E in Computer Science, GPA: 3.7

Undergraduate Coursework: Operating Systems, Databases; Algorithms, Data Structures, Advanced Computer Architecture, Web Technologies, Data Communications, Network Security. SKILLS: Java, Python, C, JavaScript, Typescript, Angular 4, HTML, CSS, Bootstrap v4, jQuery, Node.js, D3.JS, MYSQL, Linux TECHNOLOGIES: Git, IntelliJ IDEA, NetBeans, Visual Studio, Octave/MATLAB ACADEMIC PROJECTS

Question-Answering Model for Stanford Question Answering Dataset (SQuAD) (2018) (Technologies: Pytorch, Python, SpaCy) - Developed a Question Answering model which answers the questions based on the passage. The model is based on R-NET and is implemented in Pytorch. It has achieved an EM and F1 score of 75 and 85 respectively.

Tweet Generating Model (2018) (Technologies: Pytorch, Python) – Successfully implemented a tweet generating model which is trained on Donald Trump’s tweet and was able to generate tweets similar to it. GRUCell was used to achieve the best possible result.

Traffic-Sign Classification using a Convolution Neural Network (2017) (Technologies: Tensorflow, Python) – Implemented a six-layer convolutional neural network which classifies the various traffic signs into 43 classes. The model was trained on augmented data which improved the accuracy. German Traffic Sign Recognition was used for this project. The accuracy obtained on 46,000 images was 98.2%.

Decision Tree and Random Forest (2017) (Technologies: Python) – Implemented a decision tree classifier and random forest classifier consisting of 22 classes. The UCI flower dataset was used. I achieved an accuracy of 1.0.

Predictive and Statistical Analysis of Reservoir (2017) (Technologies: HTML, CSS, JavaScript, Angular 4, Node.js, MongoDB, D3.js) – Implemented a web application that successfully predicts the quantity of water in the reservoir based on the data samples given. Other features include Reservoir Map, News Blog.

Handwritten Digit Recognition (2017) (Technologies: Octave) – Implemented a Vanilla Neural Network model which successfully classifies and identifies the handwritten digit. The model was trained on MNIST dataset which has a training set of 60,000 examples, and a test set of 10,000 examples. Model achieved an accuracy of 98%

Distributed System Deadlock Analysis (2016) (Technologies: C, Linux) – Implementation of Ricart-Agrawal algorithm to handle the critical section. This was analyzed by varying the number of threads/process. 5 machines were connected to execute a task.

Smart Office Hours (2017) (Technologies: Python) – Integration of Raspberry Pi with Amazon Web Server. Implemented a system which alerts the students and the faculty about the presence of each other via communication through LEDs. MQTT protocol was used for the connection between the client and the server.

Webpage for ISA, University of Massachusetts, Lowell (2017) (HTML5, CSS, JavaScript, Angular.js)- Developed a website for the ISA of University of Massachusetts, Lowell. Link:- https://isaumasslowell.com.

Sentimental Analysis Technique using Jaccard and Cosine Implementation (2015) (Technologies: Python) – Implemented a model which performs sentimental analysis by classifying the reviews which are feed as inputs. Cosine and Jaccard similarity measure were used to evaluate the words in the review. Published paper on this topic in International Journal of Computer Application, Vol 115- No.12, April 2015. Link:- http://research.ijcaonline.org/volume115/number12/pxc3902460.pdf AREAS OF INTRESTS: Software Engineering, Data Mining, Machine Learning, Natural Language Processing