Python Assistant

Paras Gulati

917-***-**** parasgulati.com adcmq0@r.postjobfree.com

www.linkedin.com/in/paras-gulati j www.github.com/parasgulati8 EDUCATION

New York University, Brooklyn, NY Aug 2018 - May 2020 Master of Science, Computer Engineering

Concentration : Machine Learning, Deep Learning, Algorithms, Cloud Computing, Databases Guru Gobind Singh Indraprastha University, New Delhi, India Aug 2012 - May 2016 Bachelor of Technology, Electronics and Communication Engineering SKILLS

Machine Learning Classi cation, Regression, Clustering, Feature engineering, manipulation and visualisation Statistics Regression, Con dence intervals, Bayesian and Monte Carlo methods Coding Python (Pandas, Numpy, NLTK, TensorFlow, Keras, Sklearn, PyTorch), C++, C, SQL Databases Oracle, Hive, MySQL, PostgreSQL, SQL Server, MongoDB Tools Git, Shell, AWS(RDS, EC2, S3), Hadoop, Spark, Docker EXPERIENCE

Research Assistant, MERIIT Lab, New York University Sep 2019 - Present Advisor : Prof. Farokh Atashzar

Leading the AI vertical of the MERIIT lab to develop strategic plans and road-maps for research.

Developed a model for advanced prosthetic arm that classi es EMG signals of various hand gestures by using TensorFlow on HPC cloud cluster.

Proposed a novel Neural Network architecture based on Dilated CNN and LSTM that classi es 17 gestures with 86% accuracy, studied on 40 intact subjects.

Data Science Intern, Executive Business Solutions Corp, NY May 2019 - Aug 2019

Provided high-touch training to 25+ students on data visualization, regression & classi cation modeling, Bayesian statistics and time series analysis.

Designed curriculum in data science and mentored clients to develop machine learning projects such as Housing Prices Prediction, Spam Filter, Text Classi cation etc. and deployed them on AWS cloud instances. Assistant Systems Engineer, Tata Consultancy Services, India Oct 2016 - July 2018

Developed client and server applications for a broadband company to automate work

ows and add new functionalities to the billing system. Primarily utilized C++, Oracle, PL/SQL, and Unix for development.

Executed ETL operations using SQL and shell scripting. Performed data mining and data modeling on raw data using Python (numpy, pandas, scikit-learn, scipy) to estimate user demands in future. PROJECTS

Object Detection and Localization (Python, TensorFlow)

Developed an object detection system for self-driving cars that classi es objects and generates bounding boxes around them. Trained the model on GPUs in cloud cluster.

Used YOLO algorithm on 80 classes and achieved an accuracy of 82% on 40,000 test images. Anomaly Detection in Credit Card Transactions (Python, Scikit-learn)

Designed an anomaly detection system to detect fraud in credit card transactions.

Used machine learning models (Logistic Regression, Random Forest, & Boosting) to classify fraudulent transactions on a highly skewed database of 227,845 transactions. Achieved an accuracy of 94% on the test set. Emoji er (Python, TensorFlow)

Developed a NLP application which nds the most appropriate emoji to be used with an input sentence to make it more expressive. Used 50-dimensional GloVe embeddings and achieved an accuracy of 86% on test set. Jazz improvisation (Python, Keras)

Implemented a deep learning model to compose Jazz music. Trained a recurrent network using LSTM on 60 sample Jazz compositions to learn the patterns in Jazz music and generate new compositions.

Contact this candidate