Hello! This is Komal Bhimireddy 
Machine Learning Engineer
A curious builder exploring the intersection of machine learning, statistics, and real-world product systems while creating meaningful and reliable data-driven experiences.
A curious builder exploring the intersection of machine learning, statistics, and real-world product systems while creating meaningful and reliable data-driven experiences.
I completed my Master’s in Statistical Data Science at San Francisco State University, where my curiosity for machine learning, statistics, and real-world problem solving grew far beyond the classroom. During my time at Suki AI, I found myself drawn into the world of speech recognition systems, working with large-scale audio and text data to improve how clinicians interact with technology through voice. Alongside industry experience, research became another space where I learned to appreciate the depth hidden inside data, especially through my work on volatility forecasting using centuries of financial history. Beyond engineering and research, teaching calculus and statistics gave me a different kind of fulfillment, strengthening not only my technical foundation but also my ability to make complex ideas feel approachable and human.
MS in Statistical Data Science
ML Engineer Intern at Suki AI
Volatility Forecasting & Public Health Research
August 2023
May 2025
July 2025
Each project here reflects my drive to tackle challenges, implement creative solutions, and grow my technical expertise through hands-on experience.
LLM-Powered Finance Tracker
Text Analysis
Recommendation Engine
Image Processing
Applied Stochastic Models in Business and Industry
Published
In this work, we looked at over 766 years of gold return data to understand how factors like leverage, tail risks, skewness, and kurtosis influence volatility. Using Bayesian time-varying quantile regressions, we found that these moments can predict short- to medium-term volatility more accurately than traditional models. The findings, confirmed with higher-frequency data, highlight useful insights for investors and policymakers navigating uncertainty.
See Publication Details
Ongoing Study — BRFSS Data Analysis
Manuscript in Progress
This study investigates how frequent binge drinking relates to frequent mental distress (FMD) across racial and ethnic groups in the U.S., using BRFSS survey data. By controlling for demographic and socioeconomic factors, and employing logistic regression with interaction effects and complex survey weighting, the research aims to uncover disparities in alcohol-related mental health outcomes.
Kayaking
Travelling
Badminton
Music
Photography
