welcome

portfolio · projects · journey

about me

I launched my journey into tech in 2020, drawn to hard problems and continuous learning. Today I'm an AI Engineer at IBM, building generative AI, retrieval-augmented generation (RAG), and production ML workflows for enterprise clients.

I earned my M.S. in Data Science from The University of Texas at Austin in 2023. Hook 'Em! When I'm not at work, I ship side projects that mix data engineering with creativity — from systematic trading education kits to natural-language music tools — and I like tinkering with weather and geospatial data (see the radar below).

projects

featured projects

live site

open source · fintech education

Hedge Fund at Home

Educational systematic trading kit — Python services, simulated execution, Kafka & cloud patterns. Learn stat arb and production-style architecture.

live quest github

technical experience

current chapter

Artificial Intelligence Engineer

IBM · Austin, Texas · Jan 2023 – Present

  • Technical lead on 60+ generative AI projects valued at $15M for Fortune 500 and public sector clients — agentic AI, production ML, and architecture balancing inference performance with compliance.
  • Built a generative AI UI for a major oil & gas client: vector retrieval, complex filtering, pivot-style aggregations, and an evaluation pipeline combining rule-based checks with Llama 3 LLM-as-a-judge.
  • Designed an agentic AI solution for a large U.S. retailer: invoice processing (Oracle Fusion via MCP, Workday via OAS), AskHR assistant with AstraDB, and tax code classification from historical data.
  • Refactored real-time analytics pipelines for live UFC fight insights using Python, Celery distributed task queues, and Redis event streaming; implemented a plugin-based registry pattern to improve scalability across multiple Celery beat workers and developed a statistical threshold calculator analyzing historical fight data to generate predictive performance alerts with configurable win-rate correlations.
  • Architected RAG pipelines with hybrid lexical–semantic search, reranking, and embeddings (watsonx.data, Milvus, Elasticsearch) across 50k+ documents — 20% lower query latency.
  • Automated ServiceNow IT ticketing for a Fortune 500 chemical client with few-shot prompting (95% classification accuracy) and summarization layers that cut managerial review time by 25%.
  • Contributed reusable agentic AI workflows and industry assets across oil & gas, manufacturing, and supply chain teams at IBM.
  • Built Python and SQL ETL pipelines; deployed models to IBM Cloud Pak for Data with drift monitoring and governance.
earlier chapter

Data Engineer

AECOM · Austin, Texas · Dec 2020 – Nov 2022

  • Architected a geospatial database in ArcGIS Pro with normalized schema and key-based relationships across 30,000+ miles of roadway for transportation safety analysis.
  • Programmed SQL expressions in ArcGIS to query, filter, and join multi-layer spatial datasets — 40% less manual data prep for engineering teams.
  • Created interactive flood-risk visualizations with pandas and matplotlib for civil and environmental stakeholders.

education

M.S. · Data Science UT Austin · Dec 2023
B.S. · Civil & Water Resources Texas A&M · May 2020

weather · radar

Live radar + computed metrics for Austin — precipitation window, forecast sums, and hourly trend from open APIs.

Metrics refresh on load · linear trend on 6h hourly precip · educational demo

—
temp · now
—
precip · 24h sum
mm · station model
—
precip · next 6h
forecast sum
—
6h trend
hourly precip slope
—
radar frames
—
—
frame age
vs selected scan
hourly precipitation · last 24h (mm)
Loading radar…

Radar · Rain Viewer · forecast metrics · Open-Meteo · map © OpenStreetMap