Hello!
I am a T-shaped engineer currently pursuing a graduate degree at the Information School at University of Washington where I'm focusing on Data Science & User Experience. Before coming to the UW, I completed my undergraduate degree in Information & Communication Technology from Ahmedabad University. My interests lie at the intersection of data and social science, specifically using data science and learning to improve various human functions and integrate them into user-centric products.
I am interning at Daimler Trucks North America (Daimler AG) in Portland, OR as a Data Science Intern in their Manufacturing Information Systems team! I am currently responsible for building predictive models to improve manufacturing and production processes.
I occassionally play the keyboard and also like to read and explore natural language processing, specifically contextual understanding and vectors for word representation. I am in the process of making a page/blog dedicated to the same.
I am actively seeking co-op/full time opportunities in Applied Data Science/Machine Learning! Please feel free to reach out via e-mail or phone
You can download my resume here and also read my occassional musings here
Build automated systems using Alteryx, SQL to streamline production processes for DTNA. Create reporting suites to examine trends in warranty data.
Analyze text warranty data and perform clustering to categorize warranty claims.
Perform feature engineering and statistical tests between disparate data in different quality systems and develop predictive models for warranty rates based on shop floor quality results.
Build and deploy a proof of concept predictive model for warranty claim rates.
Focused on creating deep learning models using Keras for Sentiment Analysis and deploying them into the company product. Created message level models for top companies in Finance, FMCG etc. across all social channels. Improved the accuracy by ~10%.
Worked on different NLP and Machine Learning models such as TF-IDF, SVM, Naive Bayes, etc. for text classification. Researched and developed a phrase chunking component to extract noun, adjectives and verb phrases. Further improved the accuracy and precision by approximately 2%. Integral component to extract multiple sentiments.
Reduced training time by 10% by optimizing different hyperparameters in keras. Connected database through MySQL connector to give real-time sentiment analysis.
Curating data driven articles for the student organization Dubstech
Student Assistant in the HR & Payroll Team, University of Washington.
Responsible for auditing and analysis of payroll and compensation data.
GPA: 3.8
GPA: 3.0
Apart from being a full-time student, I enjoy most of my time reading fiction and playing the keyboard. In the summers, I am an amateur but avid hiker.
When forced indoors, I follow a number of sci-fi and fantasy genre movies and television shows, I love to cook, and I spend quite a bit of my free time exploring the latest developments in the natural language processing and deep learning domain.