I'm a computational engineer who designs, ships, and operates full-stack ML and LLM-based systems end-to-end. My PhD at the University of Texas at Austin grounded me in numerical methods, stochastic modeling, and scientific computing — a toolkit I now turn on quantitative finance, retrieval-augmented generation, and multimodal document intelligence.
In practice, that means systematic trading systems running in production, LLM-backed research and discovery tooling, hybrid retrieval pipelines, and the serverless cloud scaffolding around them. I care about the whole lifecycle — feature engineering, model selection, calibration, deployment, evaluation, and the boring infrastructure that keeps any of it from breaking at 3 a.m.
Two systematic trading systems run in production. The interday equity options strategy combines LoRA-tuned Chronos embeddings with per-asset TabPFN classifiers. The long-short equity portfolio is driven by LightGBM alpha/beta forecasts, genetic-algorithm optimization, and regime-adaptive rebalancing. Alongside these, I built an intraday equity and futures options strategy on anomaly-detection signals, and implemented a jump-diffusion-aware Heston calibration via constrained optimization and FFT pricing.
All of this sits on a serverless-first stack — AWS Lambda, DynamoDB, S3, EventBridge, API Gateway, CloudFormation, and Modal — wired together by GitHub Actions CI/CD that pushes digest-pinned ECR images shared across multiple bots. Lambda is tuned for ML inference with multi-GB memory and S3-backed HuggingFace and Torch caches, and React/static dashboards over the DynamoDB and S3 state are served via FastAPI + Mangum behind Lambda Function URLs.
On the LLM side, I built a Treasury and macro RAG pipeline benchmarking FAISS, Whoosh, SQLite FTS5, and Postgres tsvector retrieval; a guard-railed LLM analyst layer using Groq, LangChain, and sentence-transformers; a Llama-4-Scout multimodal PDF Q&A system with contrastive-embedding fine-tuning and Qdrant indexing fronted by a FastAPI + Next.js + Streamlit stack; an RSS-to-FX framework that turns macro news into directional LLM views scored against ground truth; a Streamlit research tool that mines rule-based trading signals against PnL via a 4-stage LLM pipeline; and a pluggable connector layer supporting 16+ tabular formats.
Underneath the production work is a steady stream of research: a structured benchmarking program across foundation time-series models — Chronos versus TimesFM, base versus LoRA-adapted Chronos, rolling embeddings over Bollinger and RSI features — for options-trade Sharpe classification; XGBoost-versus-LightGBM alpha/beta studies with SHAP and native feature importance and Kneedle-based dynamic feature selection; RSI and VIX-RSI binning studies for rule-based options entries; and experimentation with HMM volatility-regime detection, Jensen-Shannon divergence rebalancing, Black-Litterman shrinkage of ML forecasts, and NSGA-II Pareto optimization across the alpha/beta tradeoff.
I joined during the early integration phase of a surgical navigation platform combining imaging and robotic actuation. My work centered on implementing TCP/UDP communication protocols for robotic arm motion control, supporting end-to-end product packaging and deployment, and helping stage a live demonstration that anchored the company's successful Series A round.
I adapted physics-informed modeling originally developed for fusion experiment data to financial time series. The resulting production pipeline used sparse regression for PDE construction and signal generation, paired with a CAGR-maximizing Bayesian TPE optimizer driving a swing-trading system deployed on AWS Lambda.
ML & Optimization
Model development, feature engineering, and optimization across classical ML, probabilistic models, time-series foundation models, and evolutionary search.
Cloud & MLOps
Serverless-first production systems with CI/CD, infrastructure-as-code, containers, and model deployment workflows.
Programming
Python-first engineering with scientific computing, fast data processing, and frontend work for internal analytics tools.
APIs & Dashboards
Production APIs and app backends connecting market data, broker integrations, feeds, and interactive analytics interfaces.
LLM & RAG
LLM tooling and hybrid retrieval systems combining lexical and vector search for production RAG, plus multimodal document intelligence.
Beyond the code
