Selected and exploratory work
Search and filter by domain, language, and whether a project has a GitHub, demo, or paper/write-up.
Explore the work
Available links
Project index
Showing 16 / 16
Agentic financial intelligence system that reads unstructured documents, decides what matters, selectively gathers supporting context, and produces structured analyst briefs for credit-relevant events.
Ranks LLM answers by trustworthiness using single-forward-pass uncertainty signals (selective prediction).
Privacy-first Retrieval-Augmented Generation system that lets users query personal Gmail data using a fully local LLM with no external APIs or cloud inference.
Compared engineered-feature, TF-IDF, and DistilBERT models for multilabel toxic comment classification with fairness-aware evaluation.
A Streamlit-based, LLM-powered dashboard for extracting insights, detecting risks, and tracking KPIs across complex project lifecycles using document ingestion and snapshot comparison.
Built an audit-defensible system to harmonize ISO 27001, ISO 27701, and HITRUST e1 controls into a single traceable control set.
A calibrated ensemble machine learning system that estimates NFL field-goal make probabilities using play-by-play, weather, and stadium context, designed for real decision support rather than binary prediction.
Unsupervised fraud detection pipeline for Ethereum transfers that combines graph-based network features with anomaly-scoring models to surface suspicious activity at scale.
A smart job application assistant that parses unstructured job descriptions, emails, and application responses into a centralized, searchable system with AI-driven insights.
Node.js proof-of-concept that routes natural-language questions to safe, pre-defined SQL analytics functions over a DuckDB-backed dataset, then returns results in natural language using an OpenAI LLM.
Time-series-aware regression model that predicts BTC/ETH option trade prices from Binance data using engineered moneyness features and tuned GradientBoostingRegressor.
Neural network index-tracking pipeline that reconstructs S&P 500 returns using a small subset of constituents, optimizing both tracking error and portfolio efficiency under a reproducible, time-aware evaluation protocol.
A simulation-based study of how preseason rankings and perception-driven biases propagate through College Football Playoff rankings, affecting accuracy, volatility, and playoff access.
Python cost-benefit and uncertainty analysis comparing high-efficiency toilet (HET) and smart irrigation controller (SIC) rebate programs, producing per-rebate and program-level NPV metrics with Monte Carlo sensitivity modeling.
Exploratory NLP pipeline on a 20-text 2024 election corpus comparing pre-election uncertainty framing to post-election explanatory framing using TF–IDF, document similarity, syntactic complexity, and interpretable dictionary-based indices.
A custom-built Next.js portfolio platform for showcasing machine learning systems, research projects, and applied engineering work with rich filtering, featured content, and deep project write-ups.