Quantitative Researcher Machine Learning

Last updated in Mar ****

Varun Kotharkar

New Brunswick, NJ **************@*****.*** 971-***-**** linkedin.com/in/varun-kotharkar-0a792a116

Experience

Quantitative Researcher Intern, J.P. Morgan Chase – NYC June 2025 – Aug 2025 Equities Derivatives Group and Systematic Trading

• EDG Group -Developed a deep-learning engine that slashes local-stochastic volatility calibration for barrier options from hours to seconds, achieving order-of-magnitude speed-ups over traditional PDE-based solvers.

• ST Group -Engineered lightning-fast bilinear MoM estimators of power-law and Matérn roughness, turning VIX- futures data into a real-time radar for volatility regime shifts. Quantitative Researcher Intern, J.P. Morgan Chase – NYC June 2024 – Aug 2024 Market Risk Group

• Built an ML-powered VaR engine that prices risk for FX and commodity derivatives in milliseconds, capturing nonlinear tail dependencies and slashing run-times by 50 versus legacy Monte Carlo. Quantitative Researcher Intern, Van Eck / VE Labs – NYC June 2023 – Aug 2023

• Pioneered a dynamic-time-warping clustering pipeline for daily portfolio-holdings data, unearthing latent client cohorts that empowered hyper-targeted sales campaigns and drove double-digit cross-sell growth. Data Scientist / Statistician, nFerence Inc. Sept 2020 – Sept 2021

• Devised an active-learning strategy for patient labeling that slashed annotation spend and simultaneously lifted model accuracy, enabling faster iteration on large clinical datasets.

• Acted as the firm’s resident statistician, guiding cross-disciplinary ML teams on study design, bias correction, and probabilistic uncertainty—ensuring scientifically robust deliverables. Education

Rutgers University, PhD in Statistics – New Brunswick, NJ, USA Sept 2021 – present Advisors: Prof. Sijian Wang and Prof. Michael Stein

• Research: Spatial Statistics, Stochastic Processes, Machine Learning, Time Series Analysis, Deep Learning

• Proposal Defense Passed, GPA: 3.93/4.00

Johns Hopkins University, Masters in Applied Math and Statistics – Baltimore, MD, USA Sept 2018 – Dec 2019

• GPA: 3.95/4.00

• PhD Qualifying Exam passed with perfect scores

• Coursework : Matrix Analysis, Machine Learning, Measure Theory, Non-Linear Optimization Indian Statistical Institute, Bachelors in Mathematics (Hons.) – Bangalore, India July 2015 – May 2018

• First Class with Distinction, Highest Honor

• Coursework : Real and Complex Analysis, Probability Theory, Abstract Algebra, Statistics. Research

Increasing- and Fixed-Domain Asymptotics for Bilinear Estimators of a Two-Dimensional Power–Law Field Devised bilinear-filter Method-of-Moments estimators that achieve Consistent-Asymptotically-Normal (CAN) infer- ence for two-dimensional power-law fields in both fixed- and increasing-domain regimes. Robustness to irregular sampling via exact covariance-matching and matrix-operator perturbation bounds. (Submitted EJS) Extremes in Focus: A Hybrid Generative Model for Temperature Fields Designed a BATs + CVAE hybrid model that fuses Extreme-Value Theory with deep generative modeling to simulate high-resolution temperature grids, enabling physically consistent extreme event forecasting. Demonstrated improved tail accuracy and realistic spatial structure in generated climate fields. (Submitted NeurIPS) Graph Clustering via Ricci Curvature and Optimal Transport Fused Ricci-curvature weights with optimal-transport spectral clustering, proving new eigen-separation theorems that guarantee exact recovery in sparse stochastic block models and boosting community-detection accuracy on real- world networks. (In Preparation)

Disruption and recovery in physical and mental health, body mass index and smoking during the COVID-19 pandemic: a trend analysis of US BRFSS data from 2016 to 2022 Analyzed longitudinal healthcare data to understand pandemic impact and recovery. (Published BMJ Public Health) Climate-Mortality Relationships in Urban Settings

Local Climate Zone (LCZ) framework to estimate intra-urban heat stress and mortality patterns, using regression methods. (Published in Urban Climate/Sustainable Cities) Skills

Programming: Proficient with Python, R, Git, LaTeX, Matlab, SQL and Tableau Libraries: NumPy, Pandas, Scikit-learn, PyTorch, TensorFlow Teaching, Reviewing and Mentorship

• Reviewer - Annals of Applied Statistics, IEEE Transactions on Computational Social Systems, IETE Journal of Research

• Recipient, Kishore Vaigyanik Protsahan Yojana (KVPY) Fellowship — Department of Science & Technology, Government of India (2015 – 2018), a nationally competitive award supporting top 1% STEM undergraduates.

• Awarded the Kohler Scholarship, Rutgers University (1st-year PhD), a merit fellowship granted to the top cohort of incoming doctoral candidates for outstanding research potential and academic excellence.

• Teaching Assistant – Rutgers University & Johns Hopkins University: Led recitations, office hours, and review workshops for undergraduate and graduate courses in Probability, Statistics, and Linear Algebra.

• Statistical Consultant – Rutgers Office of Statistical Consulting: Advised 20 + graduate research teams on study design, advanced modeling (mixed effects, survival, spatial), and reproducibility, accelerating projects to journal submission.

• Mentor – Rutgers Data Science Club: Delivered monthly deep-dive seminars on machine-learning and applied- statistics topics.

Contact this candidate