Data Engineering · Applied AI · Cloud
Christian Bennett
I build end-to-end data platforms: automated pipelines, warehouse modeling with dbt, and live dashboards. I put AI to work at every layer, from LLM-powered product features with proper guardrails to an AI-assisted engineering workflow that ships faster without cutting corners.
Data Engineering
Automated ELT pipelines with Airflow, Python, and a dual-backend warehouse (PostgreSQL / Snowflake). Raw sources to trusted marts.
Applied AI
LLM features built with real engineering discipline: structured outputs, Pydantic validation, and prompt-injection guardrails.
Transformation & Modeling
dbt semantic layers, metrics definitions, and deduplication logic that teams can trust.
Dashboards & Cloud
Interactive Plotly + Streamlit apps served over HTTPS from AWS, deployed with GitHub Actions CI/CD.
How I work
AI is part of the toolchain, not a buzzword
AI in the loop, daily
I design, build, and review with AI coding agents as standard practice. It shortens the path from idea to shipped feature.
AI as a product feature
I built an LLM running coach into my analytics platform: it reads my training data and returns schema-validated, injury-aware recommendations.
Guardrails first
Allowlisted model inputs, strict output schemas, and cost controls, because an AI feature is only useful if you can trust what goes in and out.