Software engineer at LPL Financial, with an M.S. in Computer Science (Machine Learning) at Georgia Tech. I like problems where models meet production — pipelines, inference, NLP — and weekends behind a 35mm camera.
Kodak · Frame 10A
R1-07373
Built features for the Secured Credit Account Pilot, launching a new securities-based lending product.
Replaced a manual advisor-onboarding process with an automated HIL form for POAC, cutting turnaround time and operational overhead — then built an in-browser live editor for HIL forms that reduced testing time by 80%.
Improved FAST UI/APIs across the automated Advisor Onboarding platform, raising reliability and developer experience for the advisor lifecycle.
Architected a serverless ETL pipeline (Python Lambda, API Gateway, AWS Glue) automating nightly S3-to-Aurora parquet transfers — 2 hours down to 15 minutes, an 87% reduction.
Resolved critical bugs in the Reconciliation Breaks API, eliminating gateway errors on a high-traffic financial data pipeline, and authored BPUG documentation mapping dependencies across DIS jobs.
Led undergraduate researchers processing geospatial datasets (SRTM elevation, GIS) to reconstruct and visualize the city of Santarém, Portugal, in an interactive web model.
Authored and published a research thesis on the pipeline to the Texas A&M OAKTrust institutional repository.
A Q-Learning trading agent trained on JPM and S&P 500 history, fusing Price/SMA ratio, Bollinger Band %, and Stochastic Oscillator signals into buy / sell / hold decisions, backed by a BagLearner ensemble of Random Trees.
A KVMemNet implemented from scratch in PyTorch, using attention over key-value memory slots to answer natural-language questions about historical figures from the Wikipedia Biography dataset.
Graduation shoots, engagement sessions, and travel frames — including the one above, taken on the shore of Lake Kawaguchiko. The same patience that debugs a pipeline waits for the light.
I'm open to ML engineering and NLP-focused roles — and always happy to talk about inference optimization, trading agents, or which film stock handles snow best.