Worked under Prof. Kunal Narayan Chaudhury in Optimization problems ranging from Deep Learning to Classical Image Processing. Worked on making ML algorithms more robust, stable and generalizable.

TA for Linear Algebra and Applications under Prof. Kunal N. Chaudhury.

Optimization for Machine Learning

Was part of the Search and Discovery team working on the problem statement of finding the items that Udaan buyers are not buying from Udaan but from somewhere else. Applied many approaches ranging from simple classical linear models to Deep Learning ones. At the end what worked was a customized algorithm based on the "Missing Link Prediction Algorithm" in a bipartite graph. We modified the basic algorithm to respect the bipartite constraint. The performance was better than the baseline model that we came up. Overall the enriching learning experience and observing the dynamics of a start-up from so close definitely made me more enthusiastic to join the data science field and with some added strength in it.