Vurukonda Sathish - Ph.D.@IIT Bombay Senior Machine Learning Engineer - TensorIoT

Summary

I am a Ph.D. graduate from the Department of Electrical Engineering at IIT Bombay. My research was centered around the modeling of multivariate time series data with a particular emphasis on count data. In addition to my academic background, I have experience in AI/ML with technical leadership, strong coding skills, and a focus on effective communication and collaboration. I specialize in natural language processing and computer vision, delivering innovative strategies for business growth.

Overview

2

years of professional experience

5

Certifications

Work History

Senior Machine Learning Engineer

TensorIoT

12.2023 - Current

Anomaly detection in manufacturing data which systematically recorded from various sensors during different phases of each product's production cycle.
Developed a table detection algorithm capable of identifying tables within machine or scanned PDF documents and extracting the associated text.

Associate Research Staff Member

Vijna Labs Private Limited

03.2022 - 12.2023

Developed a document extraction product that leverages large language models (LLMs) like GPT and PALM with an impressive extraction accuracy rate of over 92%.
Developed an efficient inference engines for RNN, LSTM, GRU, and BERT model in C++ with object-oriented programming, CuBLAS, and CuDNN library and achieved an inference speed of over 70% as compared to traditional Python inference libraries.
Involved in document extraction project from from structured and unstructured documents which is based on deep learning models and transformers and achieved an impressive accuracy rate of 90%.
Built an ensemble model with Logistic Regression, Random Forest, XGBoost, Catboost, SVM, and ANN models on the extracted fields from documents to provide a conversion probability for each new submission with an impressive accuracy rate of over 93% and a regression model to prioritize new submissions.
Optimized the field extraction solution for new submissions and achieved a 70% reduction in inference time compared to the original implementation and 40% reduction in their size without any loss of accuracy after applying quantization techniques.
Developed an efficient inference engine for extracting fields from new account submissions using inference engines RNN, LSTM, GRU, and BERT model and achieved a remarkable speed improvement of 75% compared to the PyTorch implementation.
Developed an ensemble model with Logistic Regression, Random Forest, XGBoost, Catboost, SVM, and ANN models multi-model for predicting the ratings of company pros and cons based on Glassdoor reviews and achieved an impressive accuracy rate of over 93%.
Designed an ensemble of classic machine learning models to predict the confidence scores for all recognized fields on bank checks and model achieved an impressive 65% confidence score for every field on the check.

Education

Ph.D. - Electrical Engineering

Indian Institute of Technology (IIT) Bombay

02.2022

B.Tech - Electronics And Communications Engineering

Jawaharlal Nehru Technological University

Hyderabad

04.2014

Intermediate/+2 -

Sri Chaithanya Junior College

Yellandu

03.2010

Matriculation -

Zilla Parishad Secondary School

Komararam

04.2008

Skills

Softwares/Tools: LATEX, Beamer, Microsoft Office, Docker, JIRA, Git Version ControlCloud Technologies: AWS

Programming Languages: C, C (Object Oriented Programing, CuDNN, CuBLAS), R, Python (Object Oriented Programing, Numpy, Pandas, Scikit-Learn, Keras, TenserFlow, PyTorch, NLTK, Gensim, Spacy, OpenCV, Pillow, and Matplotlib), GPU Programing, MATLAB/Simulink

Platforms/OS: Windows, Linux (Ubuntu)

Academic Projects

Ph.D Thesis : Estimation and prediction over graphical models for multivariate count time series, 01/2016 to 02/2022

Guide: Prof. Debraj Chakraborty, Prof. Siuli Mukhopadhyay

Lag analysis module for an industrial die casting production line manufacturing aluminium automobile wheels, 06/2021 to 09/2021

Guide: Prof. Debraj Chakraborty

Detection model for the unknown anomalous sounds emitted from a Toy-car (ToyADMOS), 09/2021 to 11/2021
Model to predict the demand potential of products from the past details of many products, 12/2021 to 01/2022,
Course Project: Kalman Filter Design for Rotary Inverted Pendulum and Identification of State Space Model,

Guide : Prof. Debraj Chakraborty, IIT Bombay

Course Project: Design of Nonlinear Controller for Rotary Inverted Pendulum

Guide : Prof. Debraj Chakraborty, IIT Bombay

Certification

AWS Engagement Security Training for Partners

Accomplishments

All India Rank-116 in Council of Scientific and Industrial Research University Grants Commission National Eligibility Test (CSIR-UGC NET) in Engineering Science - a national level entrance examination for the award of Junior Research Fellowship (JRF), 2015.
All India Rank-8808 in Graduate Aptitude Test (GATE) in Engineering (Electronics and Communication Engineering) - a national level entrance examination for graduate study, 2015.

Publications

Journal Articles, Predictive likelihood for coherent forecasting of count time series, Siuli Mukhopadhyay and Vurukonda Sathish, International Journal of Forecasting, 38(3), 222-235, April, 2019
Journal Articles, ARMA Models for Zero Inflated Count Time Series, Vurukonda Sathish, Siuli Mukhopadhyay and Rashmi Tiwari, Statistica Neerlandica, September 2021
Journal Articles, Graphical models for multivariate count data using Monte Carlo Expectation and Maximization algorithm with `1-type regularization, Vurukonda Sathish, Debraj Chakraborty and Siuli Mukhopadhyay, Journal of Machine Learning Research, Submitted
International Conferences, ARMA Models for Zero inflated Count Time Series Modeling Dengue Incidence, Siuli Mukhopadhyay, Vurukonda Sathish and Rashmi Tiwari, International Indian Statistical Association (IISA), Mumbai, December, 2019
International Conferences, Topology selection using Monte Carlo Expectation and Maximization algorithm with `1 -type regularization for count data, Vurukonda Sathish, Debraj Chakraborty and Siuli Mukhopadhyay, Conference on Decision and Control (CDC), Nice, France, December, 2019
International Conferences, Modeling Count Time Series with Zero Inflation and Overdispersion, Vurukonda Sathish, Rashmi Tiwari and Siuli Mukhopadhyay, International Symposium on Forecasting (ISF), Thessaloniki, Greece, June, 2019
International Conferences, Predictive Likelihood for Coherent Forecasting of Count Time Series, Siuli Mukhopadhyay and Vurukonda Sathish, IIMA International Conference on Advanced Data Analysis, Business Analytics and Intelligence, Ahmedabad, April, 2019
International Conferences, ARMA Models for Zero inflated Count Time Series, Siuli Mukhopadhyay, Vurukonda Sathish and Rashmi Tiwari, Webinar on Recent Trends in Statistical Theory and Applications(WSTA), Virtual, June, 2021
International Conferences, ARMA Models for Zero inflated Count Time Series, Vurukonda Sathish, Siuli Mukhopadhyay and Rashmi Tiwari, 10th World Congress in Probability and Statistics, Virtual, July, 2021

IIT Bombay Courses

Multivariable Control Systems
Optimal Control Systems
Nonlinear Dynamical Systems
Optimization
Decentralized Control of Complex Systems
Matrix Computations
Applied Linear Algebra
Statistical Signal Analysis
Optimization in Chemical Engineering
Applied Multivariate Statistics
Advanced Process Control
Estimation and Identification

Statistical Skills and Learnings

Statistical Inference
Regression Analysis
Multivariate Statistics
Time Series Analysis
Graphical Modelling
Machine Learning
Deep Learning
Natural Language Processing (NLP)
Computer Vision (CV)
GPU Programing

Timeline