Close

Chitral Puthran

Software Engineer at Tata Consultancy Services

Download Resume

About Me

A proficient Python Programmer, specializing in Machine Learning, Data Visualization, and Data Preprocessing, with internship experience at Nokia Bell Labs in Data Science, and an active learner. A Graduate Research Assistant in Machine Learning and Big Data, with a Master of Science Degree in Computer Science from the University of New Haven, 2018. A Google Analytics Certified Individual, 2019.


In my last role, I worked at Nokia Bell Labs as a Data Science and Machine Learning Intern focused on Data Preprocessing, which involved extracting and transforming raw time-series from the World Bank, International Labour Organization and OECD websites, aggregating over 2 million data points.


I am a recent graduate from the University of New Haven, with a Master's Degree in Computer Science and Graduate Research Assistant. I have earned Google Analytics Individual Qualification, Microsoft Technology Associate and have completed several online courses on Data Science . I am a constant learner, and I am updating my skill set regularly.


Email me at chitralputhran@gmail.com
GitHub profile at github.com/chitralputhran
LinkedIn profile at linkedin.com/in/chitralputhran

Experience

Nokia Bell Labs

Data Science and Machine Learning Intern

  • Assisted the research led by Dr. Abdol Saleh at Nokia Bell Labs Consulting, by programming Jupyter Notebooks for data extraction using APIs in Python with learning instructions and reproducible code snippets.
  • Successfully collected time-series data from 1960 to 2017, for 140 macroeconomic indicators for 195 countries from The World Bank and International Labour Organization, and 256 macroeconomic indicators for 40 countries from Organisation for Economic Co-operation and Development through APIs using Python, aggregated in 2 million+ data points.

University of New Haven

Graduate Research Assistant

  • Examined various online media streaming business models, aggregated data about user ratings for movies from heterogeneous sources and forecasted ratings based on a prediction model, in a Recommender System, with the help of Python 3.6 and Apache Spark 2.2.0.
  • Developed a Tweepy Streaming Engine to collect live tweets about natural disaster situations and implemented a context filtering model to classify tweet with relevant, valuable information for disaster relief organizations and first responders.

Education

University of New Haven

August 2017 - December 2018

Master of Science in Computer Science

Coursework:

University of Mumbai

August 2013 - June 2017

Bachelor of Engineering in Computer Engineering

Relevant Courses:

Projects

Moai | Deep Learning App

This is a web application made with the help of Flask, a microframework for Python based on Werkzeug, Jinja 2, and good intentions. On the backend, a deep learning model (ResNet-50) is classifying the given image file into various categories. ResNet-50 is a convolutional neural network that is trained on more than a million images from the ImageNet database.

Project Link

Drive Curve | Machine Learning App

This is a web application made with the help of Flask, a microframework for Python based on Werkzeug, Jinja 2, and good intentions. On the backend, a Machine Learning model is used for predicting the price of the car. The machine learning model was trained on the Automobile Dataset from the UCI Machine Learning Repository.

Live Project Link

Wine Glass | Machine Learning App

This is a web application made with the help of Flask, a microframework for Python based on Werkzeug, Jinja 2, and good intentions. On the backend, a machine learning model is classifying the wine sample entered as good or bad. The machine learning model was trained on the Wine Quality Dataset from the UCI Machine Learning Repository.

Live Project Link

Amazon Alexa Reviews Analysis

The project analyzes reviews by users of Amazon’s Alexa products. Using Natural Language Processing on the product reviews and some additional features, a machine learning model should be able to predict if the feedback is positive (1) or negative (0). The primary methods used are Random Forrest and Gradient Boosting for this data-set.

View Project

Mobile Description Web Application

Designed a mobile phone database web application. Python was used to extract data of 883 mobile phone along with its features from fonoapi.

View Project

Support Vector Machine on Heart Disease Data

Applied Support Vector Machine (Supervised learning model) on Heart Disease UCI Data set from Kaggle

View Project

Tutorial for scikit-learn ColumnTransformer

ColumnTransformer was introduced in scikit-learn from version 0.20 onwards. The notebook file contains a quick and easy tutorial on ColumnTransformer to get you started.

View Project

Skills

Certifications

Get in Touch