Data Engineer with experience building scalable Data Pipelines, ETL architecture, Machine Learning in Media and E-commerce domains.
Apart from my technical interests, I share an alternate passion for music. In my spare time I enjoy hiking/outdoor adventures and aerial photography.
CGPA: 3.55
GPA: 4.0
-Deep Learning Systems
-Recognized landmarks belonging to about 15,000 classes from the largest worldwide landmarks dataset (multi-TB) with Deep CNNs and Siamese Networks using Python, TensorFlow and NVIDIA Tesla V100.
-Fine-tuned AlexNet and Inception v3 architectures with transfer learning using pre-trained ImageNet weights and achieved an accuracy of 96.88% for top 100 classes.
Won the Social Good Prize for "Hacking socio-technical challenges in our community" presented by US Ignite as part of the Reverse Pitch challenge.
GPA: 3.5
-Advanced OS
-Advanced NLP
-Exploratory Data Analysis
Built an NLTK wrapper along with WordNet’s word sense disambiguation and CoreNLP’s dependency parser to extract key linguistic features from text into a unified data object for knowledge representation in a graph database.
-Performed exploratory data analysis on Premier League data using R and ggplot to visualize relevant features, identify correlation between team attributes and increase quality of the feature space for match outcome prediction, thereby maximizing odds of winning bets.
-Attained an accuracy 6.95% higher than bookies who take bets using a Proportional Odds Logistic Regression model.
GPA: 3.43
-Algorithm Design and Analysis
-Machine Learning
-Social Media Mining
GPA: 3.67
-Elements of AI
-Advanced Database Concepts
-Music Information Processing: Audio
Graduated with Distinction
-Designed a model consisting of GPU accelerated CNNs which classifies high resolution retinal images into one of the 5 stages of Diabetic Retinopathy and an ensemble of 3 models resulted in a quadratic weighted kappa of 0.3996 on green channel images.
-Programmed the model using Python, Theano and Lasagne with the help of NVIDIA GeForce GTX 780 GPU .
Developed an online virtual stock exchange and portfolio management website using ASP.NET and Microsoft SQL Server that allows users who have registered on the website to buy and sell stocks.