Economics • Analytics • Strategy
Exploring the intersection of data, markets, and decision-making — from India to Singapore to California.
About Me
I’m an Economics major at California State University, Long Beach (graduating December 2026) with a background spanning five countries — India, UAE, Singapore, Japan, and the United States.
My work sits at the intersection of economic analysis, financial modeling, and business strategy. I’ve built revenue forecasting models that improved prediction accuracy by 20%, supported go-to-market strategies generating $1M+, and contributed valuation frameworks behind $750K ARR in enterprise deals.
Beyond analytics, I’m passionate about the beauty and skincare industry, combining deep domain knowledge of consumer trends with rigorous economic analysis. I’m particularly interested in how consumer analytics translates to investment thinking: understanding which brands are gaining share, which pricing strategies drive margin expansion, and how social media trends predict quarterly revenue before the numbers are reported.
India
Born
UAE
Early years
Singapore
2015–2021
Japan
Cultural immersion
United States
2021–present
IgotintoeconomicsbecauseIwantedtounderstandwhythingscostwhattheycost.ThatcuriositytookmefromstudyingmonetarypolicytobuildingmachinelearningmodelstotrackingskincareingredientsonReddit.Thecommonthreadisdata.Ilikefindingthepatternthateveryoneelsemissed,thenfiguringoutwhatitactuallymeans.
$0M+
GTM Revenue Supported
Market-entry strategy at Atlantic
0%
Forecast Accuracy Gain
Driver-based revenue model
$0K
ARR in Deals Analyzed
Enterprise due diligence
0
Countries
India, UAE, Singapore, Japan, US
Featured Projects
Initiated coverage on e.l.f. Beauty with an Outperform rating. Built a DCF valuation model and comparable company analysis, identifying ELF as undervalued relative to its growth trajectory and social-media-driven brand equity.
Methodology
SEC 10-K analysis → revenue build → WACC estimation → DCF with terminal value → trading comps (EV/Revenue, EV/EBITDA, P/E) → price target derivation
Key Findings
ELF trading at discount to growth-adjusted peers despite 3-year revenue CAGR of ~30%
Social media marketing efficiency 3-4x better than legacy brands on cost-per-impression basis
International expansion represents untapped ~60% revenue upside vs. current U.S.-heavy mix
Vector Autoregression model analyzing how Federal Reserve rate decisions propagate through the economy. Studies transmission to consumer spending, housing markets, and employment using 24 years of FRED macro data. Includes impulse response functions, Granger causality tests, variance decomposition, and structural break analysis around COVID-19.
Methodology
ADF stationarity tests → optimal lag selection (AIC/BIC) → VAR estimation → orthogonalized IRFs with 95% CI → Chow structural break test at 2020-Q1
Key Findings
Rate hikes transmit to consumer spending in 2–3 quarters with 89% directional accuracy
Housing starts respond 40% faster to rate changes post-2020 vs. pre-COVID
Fed Funds → CPI transmission shows 4-quarter lag with 67% of variance explained by 8 quarters
Full ML pipeline predicting consumer loan defaults on 2M+ Lending Club records. Compares logistic regression, random forest, and XGBoost with SMOTE resampling for class imbalance, hyperparameter tuning via GridSearchCV, and SHAP-based model interpretability — the same methodology used by credit risk teams at major banks.
Methodology
Feature engineering (15+ features) → SMOTE oversampling → stratified train/test → GridSearchCV tuning → ROC/AUC evaluation → SHAP feature importance
Key Findings
XGBoost achieved 0.87 AUC, outperforming logistic regression (0.79) by 12pp in minority-class recall
SHAP analysis: debt-to-income ratio and revolving utilization are top 2 default predictors
Credit score alone explains only 31% of default variance — behavioral features are more predictive
Interactive analytics dashboard tracking 25+ skincare ingredients across Google Trends, Reddit, and product launch data. SQL pipeline processes 10K+ data points into trend curves, seasonal heatmaps, and sentiment analysis — combining deep industry knowledge with data engineering.
Methodology
Google Trends API + Reddit PRAW scraping → SQLite ETL → ingredient NLP extraction → sentiment classification → Streamlit dashboard with Plotly
Key Findings
Ceramide and peptide search demand surging 140% and 95% YoY respectively
Ingredient demand signals appear 2–3 months before major brand product launches
Reddit sentiment is a leading indicator: positive-sentiment ingredients see 2.1x faster adoption
Creator Operations · Growth Strategy · Data Analysis
Tracked 50 beauty/skincare TikTok creators over 3 months to identify what drives growth. Built a creator growth playbook with phased strategies and operational recommendations for a PGC team.
Methodology
Manual tracking of 50 creators → engagement/growth metrics → cohort analysis by tier → growth playbook → ops recommendations
Key Findings
Creators posting 5-7x/week grew 3.2x faster than 1-2x/week
GRWM content had 40% higher completion rate than tutorials
Comment response within 2 hours correlated with 2.1x engagement
Simple Streamlit app that tracks monthly income, spending by category, and savings rate. Built it for myself after realizing I had no visibility into where my money was going as a student. Calculates optimal budget allocation using the 50/30/20 rule and compares against actual spending.
Methodology
CSV bank statement import → pandas category mapping → 50/30/20 rule engine → Plotly spend-vs-budget visualizations → monthly trend tracking
Key Findings
Identified $340/month in unnecessary subscriptions and food delivery spending
Improved personal savings rate from 8% to 22% within three months of tracking
Food delivery alone accounted for 18% of total spending — more than rent utilities
Scraped 15K+ Sephora product reviews to analyze what language patterns distinguish 5-star from 1-star skincare products. Used TF-IDF and basic sentiment analysis to identify which product attributes (texture, scent, packaging, results timeline) correlate most strongly with high ratings.
Methodology
BeautifulSoup scraping → text preprocessing & tokenization → TF-IDF vectorization → sentiment classification (NLTK VADER) → attribute correlation analysis
Key Findings
Products mentioning 'results within 2 weeks' received 40% more 5-star reviews
Texture complaints were the #1 predictor of 1-star ratings across all skincare categories
Packaging and 'aesthetic' mentions correlated with higher ratings independent of product efficacy
Experience
Atlantic Consulting Group
Jun – Sep 2025
Tampa, FL
RNR Marine Consultants & Engineers
Jun – Sep 2024
Singapore
Teach For India
Sep – Nov 2022
Mumbai, India
Sephora
Jul 2022 – Jan 2023
Singapore
Education
B.S. in Economics — GPA: 3.7
Expected Dec 2026
Econometrics · Financial Economics · Business Statistics · Managerial Economics
A.A. in Economics
2021 – 2025
Los Altos, California
Skills & Expertise
Technical
Analysis
Domain
Leadership
BEC — Business Economics & Entrepreneurship
Club Secretary
Organized workshops on networking fundamentals and interview readiness for professional development.
Women in STEM
Financial Advisor
Championed initiatives supporting women in STEM while delivering financial advisory services and fostering diversity.
Beyond the Resume
Thinking, Fast and Slow
Daniel Kahneman
Brandsplaining
Cunningham & Roberts
Freakonomics
Levitt & Dubner
Behavioral Economics
Skincare Science
Macro Policy
Consumer Psychology
@beautyprofessor
YouTube
@theeconomist
The Pudding
Data Viz Blog
Money Stuff
Matt Levine
Hyram
YouTube
Writing
How algorithms reshape consumer behavior and markets. Covering the behavioral economics of infinite scroll, attention as currency, information asymmetry between platforms and users, and the $24 billion influencer marketing economy.
The economics of the $670 billion beauty industry. Price discrimination from The Ordinary to La Mer, ingredient commoditization, the clean beauty premium, and how marketing creates perceived value worth more than the product itself.
The history and flaws of FICO, what machine learning models trained on Lending Club data reveal about default predictors, and the racial and socioeconomic biases baked into the system that determines who gets credit in America.
South Korea's cosmetics export strategy as deliberate industrial policy, the 10-step routine as category creation, ingredient-first branding vs. lifestyle branding, and how a country of 52 million outsells American beauty companies on their home turf.
An economics student with zero frontend experience, an AI coding assistant, and a lot of trial and error. What I learned about building things with tools I didn't understand six months ago.
Sephora stores are more operationally sophisticated than most people realize. What six months on the floor taught me about systems, cross-functional coordination, and why the best operations are invisible.
Get in Touch
Whether it’s a collaboration, a question about my work, or a conversation about economics.