May 2018 – Jul 2018 · 2 mos · Bengaluru, Karnataka, India

• Worked on pattern recognition on the surface of a ribbon to find the deformation on the surface using Deep Learning
• Used Neural Networks, Convolutional Neural Networks to model the data using Keras, Theano, Scikit Learn and similar Deep Learning Frameworks and OpenCV for Image Processing
• Under normal experimental conditions, the CNN model performed well

May 2017 – Jul 2017 · 2 mos · Bengaluru, Karnataka, India

• Detect the number of speakers in an audio using FFT, MFCC and unsupervised machine learning algorithms.
• Used Gaussian Mixture Model and K - Means Clustering to classify a sound to a particular person. We obtained greater accuracy in the GMM model than K- Means Model.
• Worked on the assessment of driving quality and road quality by collecting data from accelerometer, gyroscope and magnetometer from an android phone and designing a proper rule base.

Jan 2017 – Apr 2017 · 3 mos · IIT Kharagpur

• Designed a system to understand, identify and learn a user's set of tasks performed on a computer and find the patterns unique to the user
• Used unsupervised machine learning algorithm to group various activities of the user, minimising its time and effort
• Used Kohonen Self Organizing Feature Maps model for clustering, Powershell in Windows to gather training data for users, C++, Python to preprocess the data and Matlab for implementing Kohonen Maps Algorithm to cluster the activities of the user

Sep 2016 – Oct 2016 · 1 mo · Rajendra Prasad Hall of Residence

• We were given the datasets of the three major atm operators in the USA namely, All Point, Money Pass and CO-OP. The datasets described the different aspects of a county and the atm placed by each operator in a county. We had to find that which of the providers had placed its atms most optimally. We used two different decision-making schemes. Finally we drew the inference about which atm service provider had distributed its atms most optimally.

Jul 2016 – Apr 2017 · 9 mos · IIT Kharagpur

• I was responsible for running of the mess of RP hall having 750 boarders with my mess committee members. Along with proper running we brought it some important changes for the mess by painting and renovating the mess.

Jul 2016 – Nov 2016 · 4 mos · IIT Kharagpur

• We were responsible for teaching and clearing doubts to the first year undergraduate students who were enrolled in the programming and data structures course in the autumn semester.