Skills
What I use to move from a manual process to a working system
The tools matter. So do the choices about evidence, safety, ownership, and what should remain a human decision.
AI systems
I design workflows where an LLM uses tools, follows project instructions, and stops when a person needs to decide.
- LLM pipeline design
- Evaluation design
- Claude Code skills
- MCP integrations
- Faithfulness checks
- Human-in-the-loop workflows
Engineering & data
I connect systems, structure recurring work, and build the checks needed to operate it safely.
- Python
- API integrations
- SQL & BigQuery
- Alteryx
- Data validation
- Automated testing
- Git & versioned handoffs
Research & analytics
My research training shapes how I define evidence, compare judgments, and investigate surprising results.
- Study design
- Inter-rater reliability
- Statistical analysis
- Classical machine learning
- Experimental validation
- Physiological data analysis
Delivery
I translate a process into something another person can understand, run, and maintain.
- Requirements discovery
- Process mapping
- Runbooks & documentation
- Stakeholder communication
- Operator training
- Technical handoff
See the skills in context
The project pages show how these capabilities work together on real operating problems.
Explore projects