Art Turner
AI Engineer · Python · Cloud ML · Deployment
Data-driven by default. Relentless by nature.
- • Build and deploy ML + LLM solutions end-to-end: data → model → evaluation → delivery.
- • Strong Python/SQL foundation with a production mindset (pipelines, reliability, automation).
- • Portfolio spans NLP, time-series forecasting, and AI-enabled workflow automation.
Projects
Selected work (problem → approach → result).
Branching Scenarios
Branching-narrative learning app powered by a JSON state-machine engine.
- JSON-defined decision trees with variable tracking + conditional logic
- Google Sheets data capture for completion + reflection responses
- Multi-scenario content authoring without code changes
PythonStreamlitJSONPandasGoogle Sheets APIOAuth2Git/GitHub
StudyTok Pipeline
Pipeline that turns transcript libraries into validated quiz banks using an LLM.
- Multi-stage ETL with schemas/validation and checkpointed resumability
- Caching/checkpointing reduces rerun costs and improves fault tolerance
- Exports to Anki + mobile JSON; deployed as a responsive web app
PythonPydanticClaude APICLIAsyncNext.js (static)Vercel
Interactive Video Quiz Platform
Transforms video into stop-and-check learning with timed embedded questions.
- Pauses video at timestamps to collect comprehension checks in real time
- Dual persistence paths (Apps Script + Netlify Functions) into Google Sheets
- Roster autocomplete + accessible responsive UI
JavaScript (ES6)HTML/CSSNetlify FunctionsGoogle SheetsCloudflare R2
Watt Wise
End-to-end time-series forecasting app for building energy consumption.
- ARIMA/SARIMA baselines + ML/DL comparisons
- Feature engineering with exogenous variables
- Streamlit dashboard for interactive predictions
Pythonpandasstatsmodelsscikit-learnXGBoostTensorFlow/KerasStreamlit
Skills
Languages
Python, SQL, TypeScript/JavaScript
ML / AI
scikit-learn, PyTorch, forecasting, NLP/LLMs, evaluation, feature engineering
Cloud / Delivery
AWS, Docker, Git, CI/CD, APIs (FastAPI), Streamlit/Gradio