Master's in Electrical and Computer Engineering, Purdue University | Bachelors in Instrumentation and Control Engineering, NIT Tiruchirappalli
About
Mankind’s first brush with technology came with the use of fire and stone tools that enabled humans to survive predators and thrive. However, in recent years, technology has a reputation of being a cold and an unemotional instrument that serves to perpetuate economic interests over human well being. I firmly believe that technology acts as a powerful tool to alleviate suffering and pain in the hands of a humane and an empathetic individual.
I'm Venkatesh, pursuing my Master's in Electrical and Computer Engineering in Purdue University. Over the past three years, I've worked on projects pertaining to Signal Processing, Computer Vision and Artificial Intelligence. Owing to my inherent nature of being an empathetic individual, I have pursued many projects pertaining to medical care where I've utilized my prowess in Signal Processing. Nevertheless, I'm also adept in utilizing my skills in Signal Processing in various domains such as Robotics, Circuit Design etc. Apart from tech, I'm also passionate about cricket, carrom and am an avid writer on movies, career guidance and technology.
​
I am actively seeking full-time roles in the field of Signal Processing & Machine Learning starting Summer'21. Kindly contact me via sriniv91@purdue.edu / venkateshbs.71097@gmail.com for discussions and letting me know of any possible opportunities pertaining to my profile!
Skills
Programming Languages: Python, C/C++, Embedded C, MATLAB, Simulink, SQL.
Frameworks & Packages: PyTorch, Keras, Tensorflow, AWS SageMaker/Athena, scikit-learn, OpenCV, Git.
ML/DL techniques: CNN (ResNet, VGGNet, CapsuleNet, MobileNet etc.), RNN, PCA, GAN, SVM & KNN, Deep Generative Models (VAE, Normalizing Flows), Clustering, Logistic Regression.
Operating Systems: Windows, Ubuntu
Work experience
Click each block for detailed description
Projects
Machine Learning / Deep Learning implementations
ResNet
InceptionNet
GoogleNet
MobileNet
Semantic Segmentation
Object Detection
LSTM
GAN
Autoencoder
Real-time Emotion Recognition via facial expressions
Spider R&D annual exhibit (2018)
In order to predict the emotional state of a person via facial expressions, we used a Facial Emotional Recognition (FER 2013) dataset as the training data. Convolutional Neural Network (LeNet) was implemented to build the model with an accuracy of 96%. Python was used, where the network was built using Keras in Tensorflow backend.
Sound-zoned -Â Separating the multiple audio channels
GitHub repo with code & report
Independent Component Analysis (FASTICA Algorithm) concept was used in order to separate mixed multi-channel audio signals into individual components. The deconvoluted signals, which represent the numerous individual signals from the mixed multi-channel audio signals were plotted using MATLAB and analyzed.
Simulation of Variable Stiffness Actuator
Research work under Prof. Dhanalakshmi
The aim was to design a robotic arm capable of linear & angular motion using a bidirectional shape memory alloy. So, the simulation was done as the first stage where 180° rotation of the arm was enabled using a spring in ANSYS.
Non-Invasive Glucometer using Near Infrared Spectroscopy
Shaastra Circuit Design Challenge - 3rd place
As an alternative to the existing invasive technique of Glucose measurement, we went in for a non-invasive type of Near Infrared Spectroscopic technique where we designed a circuit to analyze the intensity of signal incident on the finger using IR LED and Photodiode. Then, we Collected the signals and predicted the approximate sugar level using Machine Learning.
A Personalized Intelligent Monitoring System for Rescue Missions*
Honeywell India Hackathon - 1st place
The main purpose of the project was to track the humans and recognize their activities in real-time without GPS via Machine Learning where the position, as well as the orientation of motion, is obtained by processing the accelerometer, and magnetometer values got from the IMU and print the map in a mobile app. This module also consisted of a health pack to monitor one's heart-rate, temperature etc.
*IEEE ICCCI conference, 2019 publication.
​
Contact
Interested in learning more about me, my work or how we can collaborate on an upcoming project? Feel free to reach out anytime, I would be more than happy to chat.