Hi there! Welcome to my digital garden.

During my 15-month internship at HP Inc., I dove into machine learning forecasting, tackling challenges from SKU-level predictions to data management. Collaborating with the SPaM team, utilizing innovative tools, and embracing HP’s culture of innovation and failure, I emerged with invaluable skills, insights, and memories.

My notes on talks I attended (mostly on LLMs) at 29th ACM SIGKDD 2023 at Long Beach, CA

This piece explores the intricacies of coffee, from the influence of its origin and roasting process to the importance of freshness and brewing techniques, primarily from my experience.

A new playlist of random songs that I haven’t listened in a while generated automatically every Monday

Creating ML demand forecast for print products at HP Inc. using LightGBM and pushing it to production for wide adoption.

Predicting race, age and gender of faces using Variational Autoencoders and Convolutional Neural Networks.

We proposed a novel end-to-end approach, the Neural Lagrangian Selling (NLS) model, to improve Guaranteed Delivery (GD) advertising by concurrently predicting ad impression inventory and optimizing contract allocation

In this lab session, I share how to use apriori algorithm for association mining. The goal is to find useful causal and association rules which can help in designing promotions for the company. Plus, you get to see what’s served at an Indian cafe.

Support vector machines (SVM) are remarkable for the unification of geometric theory, elegant mathematics, theoretical guarantees with strong practical use cases. In this blog post, I demonstrate certain properties of SVM and how to use them with caret package in R.