Kavya Agar
Computer Science @ Texas A&M University
> Building systems · Fueled by espresso · Shipping clean code
scroll to explore ↓
[ ABOUT ME ]
The Portfolio
Kavya Agar
CS Student · Software Engineer · ML Enthusiast
CS student at Texas A&M University working at the intersection of software engineering, machine learning, and cloud infrastructure. I care about building reliable, production-ready systems — from fine-tuning LLMs to shipping full-stack applications with real users.
[ MARKET DATA ]
[ EDUCATION ]
Texas A&M University
B.S. in Computer Science · Minors: Statistics, Business
Aug 2024 – May 2027
[ SKILLS & TECHNOLOGIES ]
Tech Stack
Proficiency measured in market confidence
LANGUAGES
FRAMEWORKS & ML
CLOUD & DEVOPS
Also familiar with:
[ WORK EXPERIENCE ]
The Ledger
Jun 2023Aug 2023
Software Engineering Intern · Team Lead
JPMorgan Chase & Co.
Houston, TX
- ▸Engineered a multithreaded Java data pipeline for a distributed system migrating 10TB+ from Apache Cassandra to AWS S3, implementing custom Parquet conversion and improving migration efficiency by 90%
- ▸Collaborated with 3 senior engineers through 10+ code reviews to harden system reliability, performance, and data integrity, writing unit tests and validating end-to-end pipeline correctness
- ▸Led daily Agile stand-ups for an 8-person cross-functional team, coordinating sprint execution and presenting the final solution to Executive Directors
Jan 2024Apr 2024
Technical Business Analyst Intern
Aspen Technology
Houston, TX
- ▸Built automated Python web-scraping pipelines using BeautifulSoup and Selenium to analyze 5+ competing analytics platforms, reducing manual research time by 60%
- ▸Partnered with PMs and engineers to translate data insights into product strategy decisions, contributing to an estimated 15% revenue impact
[ PROJECTS ]
The Portfolio
6 projects shipped · 3 in production · ∞ ideas in the pipeline
Financial NER Analyzer
LIVEProduction-ready document classifier with custom entity recognition for transaction amounts, dates, and account numbers. Fine-tuned DistilBERT on 116K+ tokens achieving 98% accuracy.
Stock Sentiment Dashboard
LIVEEnd-to-end ML pipeline serving 50+ users with real-time sentiment analysis from Reddit and Finnhub. Ensemble NLP inference achieving 78% accuracy with CloudWatch health monitoring.
FIN-ance
LIVECross-platform payment and expense-tracking app with ML-based spending forecasts (90% accuracy). Django REST API handling 1,000+ daily requests with OAuth 2.0 and iOS WidgetKit client.
Quant-Options-Engine
LIVEFrom-scratch options pricing engine implementing Black-Scholes Greeks, 50K-path Monte Carlo GBM simulation with antithetic variance reduction, and a Brent's method IV solver across live options chains via yfinance/FRED. FastAPI backend serving a React dashboard with real-time Greeks sliders and a theoretical vs. market mispricing chart.
Portfolio V3
WIPThis site — finance terminal aesthetic with live market ticker, animated chart, command palette (⌘K), and real-time S&P/DOW/NASDAQ data.
Pomodoro
OPEN SOURCEMinimalist focus timer implementing the Pomodoro technique with session tracking and customizable work/break intervals.
// live data
S&P 500 Heatmap
Real-time market performance by sector — block size by market cap, color by % change today.
[ CONTACT ]
Open a Position
Place your order — I respond within 24 hours
[ ORDER BOOK ]
“The stock market is a device for transferring money from the impatient to the patient.”
— Warren Buffett