Data & AI Engineer in Houston building production systems — the kind with rate limiters, kill switches, and a 3 a.m. pager, not the kind that lives in a notebook.
Day job: I own an end-to-end Toast POS → PySpark → Delta Lake → Power BI pipeline serving 25 franchise locations and millions in tracked revenue. Nights: I ship full-stack AI — multi-model LLM gateways, hybrid RAG with reranking, agentic loops, and vector search — all of it live in production on real traffic.
The overlap is the whole point. Most people doing AI can't build the data layer. Most people building data layers can't ship AI. I do both.
📍 Richmond, TX · 🎓 BS Computer Science, UT Dallas · 📜 Databricks Certified Data Engineer Associate 🟢 Open to Senior Data Engineer / AI Engineer roles — remote or Houston hybrid
|
Full-stack AI platform · Node.js · Vercel · Upstash Not a portfolio site — an AI application that happens to have my resume in it.
|
Python ETL · Vercel · Toast API Operational intelligence for 25 Nothing Bundt Cakes stores across TX / NJ / CO.
|
|
LLM resume intelligence · Python · Docker · Railway Parses resumes into structured candidate intelligence and scores fit against a live job description. Containerized, custom domain, deployed and indexed. Three-runtime distributed system Generates personalized running routes from your Strava history, calibrates distance against the OpenRouteService API with a bounded proportional retry loop, and pushes the course directly to a Garmin watch — including a sideloaded Connect IQ app I built for the FR970. |
Mistral-7B · QLoRA · MLflow · Unity Catalog End-to-end fine-tune of a PySpark coding assistant. Hybrid architecture spanning Databricks (CPU prep) and Colab T4 (GPU train), with MLflow experiment tracking and LoRA adapters shipped to HuggingFace Hub. Zero cloud dependency A ReAct agent built from scratch against a local Ollama model — no API, no network. Built to understand what LangGraph is actually doing under the hood before trusting it in production. |
AI / LLM Engineering
RAG (hybrid + RRF + reranking) · CRAG · Agentic loops / ReAct · QLoRA fine-tuning · MLflow · Evals · Prompt engineering · Vector search (Upstash, FAISS, Chroma)
Data Engineering
Medallion architecture · Incremental ETL with control tables · REST API ingestion · Idempotent replay · Data quality & reconciliation · Dimensional modeling
Platform & Infra
Marathoner chasing sub-3:00 — which is why half my side projects are running infrastructure. Also: aviation, weather systems, NBA, independent science, and the kind of open-ended research rabbit holes that end at 2 a.m. with a new repo.
Building AI that ships. Open to what's next.
