Medical chronology generator
An LLM pipeline that turns raw, messy medical records into clean, chronological case summaries for legal teams — collapsing hours of manual review into a fast, structured draft.
Senior Programmer Analyst · Data & AI Engineer · Houston, TX
I build data & AI systems teams actually rely on.
I'm a data engineer with a decade of turning scattered, messy data into pipelines, warehouses, and dashboards that large organizations — a 5,000-user school district, legal firms, analytics teams — use to make real decisions.
Lately that's grown into AI: I've built and shipped machine-learning and LLM systems — a medical-record summarizer, an automated essay grader, a lead-scoring model — using Python, PyTorch, TensorFlow, and the OpenAI API.
AI & machine learning
Automated essay scoring
A machine-learning model that grades student essays consistently and at scale, giving educators faster, more uniform feedback across the district.
Lead scoring & prioritization
A predictive model that ranks incoming leads by conversion likelihood and urgency, so intake teams spend their time on the highest-value contacts first.
Selected work
30%
faster data retrieval — consolidated 50+ sources into a single warehouse supporting company-wide business intelligence.
75%
less processing time — rebuilt legal-data pipelines with Pandas & Kafka across Filevine, Lead Docket, SmartAdvocate, and more.
65%
more analysis efficiency — engineered a scalable AWS + PostgreSQL warehousing solution from the ground up.
40%
lower technology spend — designed custom tools and integrations that replaced costly third-party platforms.
What I work with
Data & infrastructure
- Python
- PostgreSQL
- AWS
- ETL / Data Warehousing
- Kafka
- Pandas
- Docker
- Tableau
- Looker Studio
AI & machine learning
- PyTorch
- TensorFlow
- scikit-learn
- OpenAI API
- LLMs
- Custom model training