Senior Associate Algorithm Engineer at QuinStreet.
I build production machine learning ranking systems, conversion models, and marketplace optimization engines.
Currently owning the Personal Loans ML ranking architecture at QuinStreet.
I’m a Senior Associate Algorithm Engineer at QuinStreet with 5+ years of experience building
production machine learning ranking systems in AdTech and data engineering. I own the end-to-end ranking architecture for the
Personal Loans vertical across 3M+ annual inquiries.
My work spans CTR and conversion modeling, constrained marketplace optimization, large-scale Cython simulation, and online experimentation.
Before QuinStreet, I interned in Machine Learning at VidMob and worked as a Software Engineer at Accenture. I earned a
master’s in Electrical and Computer Engineering from UCLA and a bachelor’s in Electronics and Communication Engineering
from PES University, and have published research in computational imaging at SIGGRAPH 2022.
Senior Associate Algorithm Engineer · Personal Loans Ranking
July 2026 – Present
Technical owner of Personal Loans ML ranking for 3M+ annual inquiries, driving expansion from subprime to higher-quality traffic (~$2M margin opportunity).
Built the Personal Loans XGBoost vendor conversion model and own the CTR model with weekly cron retraining; a persistent 90/10 holdout shows about a 10% profit lift on a $1M+ margin segment.
Lead a margin preservation initiative combining CTR and CVR predictions to filter unprofitable leads before purchase, projecting an 8–12% reduction in wasted lead spend.
Own the Cython simulation framework for pre-release stress testing and daily Auto Insurance production health monitoring, cutting runtime by 50%; mentor engineers and authored team-wide model release standards.
Built and shipped production XGBoost CTR and conversion ranking models across Personal Loans, Auto Insurance, and Banking, averaging a 10% improvement over legacy lookup baselines.
Implemented the Hungarian algorithm for optimal constrained ranking assignment, validated through staged online A/B tests to lift margin revenue 2.5–8% across Auto Insurance and Banking.
Optimized the Cython simulation suite to 95% production fidelity and engineered high-impact features guided by SHAP attribution and permutation importance.
Software Engineer · Healthcare & Pharma Data Engineering
June 2019 – July 2021
Automated pharma sales ETL workflows in Python, SQL, and Apache Hive, cutting analysis time by 50%.
Built drug-sales analytics and Tableau dashboards, lifting competitor match rates by 20%.
Onboarded 10 graduate engineers and was named a Global Innovation Challenge Semifinalist.
Research experience
Student Researcher · UCLA Visual Machines Group
July 2020 – January 2023 · Advised by Prof. Achuta Kadambi
Diagnosed skin-tone bias in remote health sensing and co-authored the SIGGRAPH 2022 paper blending 77 GHz radar and RGB camera sensing, improving accuracy by 75%; developed a shift-robust rPPG loss function that reduced error by 40% and wrote 2 chapters for an MIT Press Computational Imaging book.
Teaching Associate · UCLA
March 2022 – June 2023 · Los Angeles, California
Led 2 discussion sections and 2 lab sessions guiding students in mathematical modeling in Python, and supported faculty with course preparation and grading.
Research Intern · Indian Institute of Science (IISc)
January – June 2019 · Advised by Dr. Suresh Sundaram
Restructured the ABIDE fMRI dataset using inter-regional brain correlation analysis and trained deep neural networks for early Autism detection, achieving 80% accuracy (GitHub).
03. Selected work
A few projects I’ve led across marketplace ML ranking, constrained optimization, and computational imaging. Browse the archive for additional projects and research.
FEATURED · 01
Personal Loans ML ranking & margin preservation
Owning the end-to-end CTR, conversion (CVR), and lead-filtering ML ranking architecture across 3M+ annual Personal Loans inquiries at QuinStreet.
The challenge
Legacy lookup-table baselines struggled to capture borrower intent and publisher traffic quality across 3M+ annual inquiries, leaving margin untapped and spending on low-conversion leads.
My contribution
Built the Personal Loans XGBoost vendor conversion model and own the CTR model with weekly cron retraining, SHAP-guided feature engineering, and a joint CTR and CVR lead-filtering system.
Replacing greedy offer allocation with optimal constrained assignment (Hungarian algorithm) backed by a 95%-fidelity Cython simulation testbed.
The challenge
Match consumer intent to advertiser offers under eligibility and budget constraints while stress-testing ranking policies offline at production parity before release.
My contribution
Implemented the Hungarian algorithm for optimal constrained ranking assignment and re-engineered the offline simulation testbed in Cython with automated daily cron health monitoring.
Blending RGB camera video with 77 GHz millimeter-wave radar and a shift-robust loss function to mitigate skin-tone bias in remote heart-rate sensing.
The challenge
Camera-based remote photoplethysmography (rPPG) degrades on darker skin tones due to optical absorption, and unsynchronized camera-oximeter capture introduces temporal lag.
My contribution
Contributed to an adversarial multimodal fusion pipeline combining 77 GHz radar and RGB video at UCLA VMG, and developed a shift-robust PyTorch loss function for temporally misaligned PPG signals.
Machine Learning Intern · June 2022 – September 2022
Designed deep learning computer vision models for object detection and creative ad asset scoring in PyTorch, increasing ad engagement classification accuracy by 25% across automated recommendation pipelines.
PyTorchComputer visionAd engagementA/B testing
IISc BENGALURU
Autism detection from fMRI scans
AI & Robotics Lab · January 2019 – June 2019
Restructured the ABIDE neuroimaging dataset using inter-regional brain correlation analysis and trained deep neural networks for early Autism Spectrum Disorder detection, achieving 80% accuracy.
Deep learningfMRI neuroimagingABIDE dataset
ACCENTURE
Pharma ETL & sales analytics
Healthcare & Pharma · June 2019 – July 2021
Automated pharma sales ETL pipelines in Python, SQL, and Apache Hive to cut analysis time by 50%, and built Tableau analytics dashboards that lifted competitor match rates by 20%.
