About Me

Raman Ahuja

I am Computer Science graduate student at ASU, currently focused on Machine Learning.
Previously, I worked as a Software Engineer at Numerify- System of Intelligence for IT, where I developed optimized engine for large-scale distributed columnar databases. I have experience of delivering production ready solutions.
I am a design oriented person and focused on implementing efficient algorithms to develop end-to-end products.
I love reading and I keep my self up to date with the emerging technologies.
I am currently seeking internship opportunities for Summer 2019.

My Career

Arizona State University - FALL 2018

My interest focuses on Machine Learning and Deep Learning.
Subjects of Interest - Statstical Learning and Pattern Recognition, Topics in NLP, Foundations of Algorithms

August. 2018 - May.2020
Computer Science Graduate Student
ASU

Numerify.Inc

I primarily contributed on three projects -

  • Developed optimization platform for large-scale columnar databases. Engine is deployed for 30+ tenants and reduced daily ETL batch processing time from ~10 hrs to ~6hrs.
  • Enterprise social media feedback driven analytics platform, led to 20% faster resolution of incident tickets.
  • Healthcare outage analytics application, elevated outage prevention by ~2x.

Tools & Skills - Java, Python, AWS Redshift, AWS, Mysql, ETL, Talend, Microstrategy, scikit-learn, nltk .

Jun. 2016 - Jun. 2018
Associate Software Engineer
NUMERIFY

Visvesvaraya National Institute of Technology, Nagpur

I completed my Undergraduation Thesis on - Enhanced Naïve Bayes model in Natural Language Processing for Sentiment Analysis, which can be generalized to a number of text categorization problems for improving speed and accuracy.

Jun. 2012 - May. 2016
B.Tech Computer Science
VNIT

Doboz

Designed and developed an android application for digital wallet as a service. Implemented loyalty points program, rewards and incentives, and gift cards purchase-sale on the platform.

Tools & Skills - Java, Android, RDBMS

May. 2015 - Jul. 2015
Software Engineering Intern
DOBOZ

Centre of Excellence Lab - VNIT

Designed and developed a device (hardware and software) for curing partial colour blindness, using optimal wavelength for better response of retina. Research and Development was fully funded under (TEQIP- II) program.

May. 2014 - Jul. 2014
Research Engineering Intern
VNIT

My Skills

My Projects

Enhanced Naive Bayes Model in NLP for Sentiment Analysis

Developed a novel approach for improving the accuracy of Naive Bayes classifier for sentiment analysis. Factored in methods like effective negation handling, word n-grams and feature selection by mutual information and observed amalgamation of these results in a significant improvement in accuracy. We achieved an accuracy of 85.0% on the popular IMDB movie reviews dataset. Thus built a highly accurate and fast sentiment classifier which has linear time training and testing time complexities which is at par to the start-of-the-art SVM models. The proposed method can be generalized to a number of text categorization problems for improving speed and accuracy.
Tools & Implementation - Java, StanfordCoreNLP, Text Pre-Processing, Feature Engineering, Feature Extraction.
Dataset - Stanford Large Moview Review Dataset

...   ...

Intruder Detection through Webpage Session Tracking

Goal is to identify a user on the Internet tracking his/her sequence of attended Web pages. The algorithm built will take a webpage session (a sequence of webpages attended consequently by the same person) and predict whether it belongs to genuine person or somebody else. Have done feature-engineering based on sequential pattern mining techniques to enhance ROC AUC.
Tools - Numpy, Pandas, scikit-learn, matplotlib.
Dataset - Blaise Pascal University proxy servers

Machine Learning Coursera,Stanford University - Prof.AndrewNg

Completed graduate level course with 97.30% and developed detailed solutions in MATLAB. Completed projects based on Linear Regression, Logistic Regression, Multi-class Classification and Neural Networks, SVM, PCA, Recommender Systems.

...   ...

Device for Curing Color Blindness - COE Lab, VNIT

Designed and developed a device (hardware and software) for curing partial colour blindness, using optimal wavelength for better response of retina. Research and Development was fully funded under (TEQIP- II) program.