Experience

 
 
 
 
 
May 2019 – August 2019
Malvern, PA

Machine Learning Engineer

Vanguard


  • Implemented distributed model training on AWS-based Spark cluster to parallelize training of Deep Learning models
  • Developed modules for training Vanguard specific GloVe embeddings. Achieved 10x processing speedup using Cython and Dask
  • Carried out extensive exploration of a web-interaction dataset, to engineer new features and derive actionable insights
 
 
 
 
 
August 2018 – Present
Philadelphia, PA

Deep Learning Programmer

Computational Breast Imaging Group


  • Develop Deep Learning frameworks for prediction of Breast Cancer using longitudinal patient mammogram data
  • Designed CNN and Siamese Network models using Transfer Learning in PyTorch framework
 
 
 
 
 
May 2016 – June 2018
Gurgaon, India

Business Analyst

American Express


  • Developed and maintained credit risk assessment models for all non-US markets covering 12M Amex customers
  • Primary developer of AAT, an automated market-specific model adjustment tool utilizing SAS and Shell scripting
  • Developed adjustments for UK CDSS and TSR models in Q3’16 and Q2’17, with benefits of $2M and $5M respectively
 
 
 
 
 
February 2016 – April 2018
Gurgaon, India

Data Scientist

Freelancer.com


  • Provided Machine learning analysis/consultation on 20+ projects for clients from US, Germany and Australia
 
 
 
 
 
January 2015 – May 2015
Chennai, India

Intern

System Insights


Worked on prediction of quality parameters for 3D printing process based on 3D printing process variables

Roles

Jul 2019

Lead, ML Screening Team

Jarvis


  • Helped in the development of the screening process for onboarding new freelancers
  • Conduct interviews to screen freelancers and evaluate their skillset and area of expertise
Feb 2019

President

Penn Data Science Group


  • Conceptualized an Outreach team to help establish corporate relationships
  • Sourced projects from ACLU, Penn Baseball, Teach For America to provide industry experience for club members
  • Conducted workshops on Introduction to AWS, R Shiny Development and Distributed Learning using AWS
Jan 2019

Teaching Assistant

University of Pennsylvania


  • Teaching assistant for CIS 519 : Applied Machine Learning
  • Teaching assistant for CIS 545 : Big Data Analytics
Jul 2018 – Dec 2018

Project Analyst

Wharton Analytics Fellow


  • Work with Citi Ventures to develop Machine learning models to predict customer churn rates
Jan 2016 – Dec 2017

Course Mentor

Coursera


  • Mentored 2 courses, offered by the University of Washington as part of the Machine Learning specialization
  • Moderate discussion forums for clarification of doubts. Help refine the course content to improve learning experience
May 2015 – Jun 2016

Head

Analytics Club, IIT Madras


  • Lead a team of 6 to manage sessions for 350+ club members. Currently supervising 20 students working on 4 projects
  • Conceptualized and organized a Summer School on data analytics and machine learning for 40+ students

Projects

*
All Deep Learning Computer Vision NLP

Deep Learning for Chest Xray Diagnosis

Detection of 13 different pathologies using Deep CNN models

Deep Learning methods for classification with Limited Datasets

Explored the use of Siamese Networks for 2 classification tasks - Digit and Face Recognition

Design of Neural Networks using NEAT

Design of optimal neural net architecture for game-playing bot

Optimization of Bayesian Networks

Genetic algorithms for optimizing Bayesian network architecture for classification tasks

TicTacToe Bot

Python app implementing Q-Learning based TicTacToe playing bot

Training Domain specific Glove Embeddings

Evaluated the performance of domain specific Glove vectors trained on relevant data and its comparison with pretained Glove vectors

Tumor Detection in Lymph nodes

Detection of Breast Cancer by analyzing lymph node tissue images

YOLO

Implemented YOLO from scratch in Keras for basic objection detection tasks

Brain Tumor Segmentation using Conditional Random Fields

Master’s Thesis on Brain Tumor segmentation on BRATS 2015 data using a combined Stacked Denoising Autoencoder - CRF model

Water Pump Dashboard

RShiny app to visualize the water pumps in Tanzania and run predictive model

Courses
Introduction to Big Data with Spark Fast.AI Scalable Machine Learning Time Series Analysis
Big Data Analytics (Advanced Track) Artificial Intelligence Statistical Learning Machine Learning
Biostatistics Data Structures and Algorithms (Python) Analytics Edge Practical Predictive Analytics

Skills

Python

R

SAS

Keras

Tensorflow

PyTorch

PySpark

SQL

Java

Linux

C / C++

D3.js

Contact