Hello! This is Komal Bhimireddy 
Data Enthusiast
A curious mind navigating through the exciting world of data, blending statistical expertise with machine learning to bridge the gap between raw data and impactful decisions.
A curious mind navigating through the exciting world of data, blending statistical expertise with machine learning to bridge the gap between raw data and impactful decisions.
I graduated with a Master’s in Statistical Data Science from San Francisco State University, where I built a strong foundation in statistics, machine learning, and data analysis. Recently, I interned at Suki AI, where I designed and fine-tuned ASR pipelines using Suki's ML Platform, worked with large-scale audio/text datasets, and improved Whisper’s performance for real-world clinical workflows. Alongside industry experience, my research on volatility forecasting has been published in a peer-reviewed journal, giving me the chance to apply advanced statistical methods to century-spanning data. Beyond academics and work, I’ve enjoyed teaching as a Graduate Teaching Associate and Math Tutor, helping students navigate challenging concepts in calculus and statistics. These experiences not only sharpened my technical skills but also strengthened my ability to explain complex ideas simply - something I value both inside and outside of data science.
MS in Statistical Data Science
ML Engineer Intern at Suki AI
Volatility Forecasting & Public Health Research
Each project here reflects my drive to tackle challenges, implement creative solutions, and grow my technical expertise through hands-on experience.
Text Analysis
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
Work 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
