How I work
Practice notes on the three kinds of work I do: building agent systems, teaching teams to build with AI, and fixing the engineering that gets in the way. Written down so the method is as legible as the output.
AI pipelines that do real work, not just demos.
- End-to-end agent pipeline architecture and implementation
- Tool and API integration for autonomous agent actions
- Error recovery, guardrails, and human-in-the-loop fallbacks
- Evaluation frameworks to measure agent reliability over time
Read the notes →
Get your team building with AI instead of just talking about it.
- Prompt engineering and LLM integration patterns
- Agentic design patterns and multi-step workflows
- Evaluation, testing, and reliability strategies for AI systems
- Tool-assisted coding workflows with AI pair programming
Read the notes →
I fix the systems everyone else is afraid to touch.
- System diagnosis and root-cause analysis for critical issues
- Performance optimization and architecture remediation
- Greenfield architecture and implementation for new systems
- Technical debt reduction and codebase modernization
Read the notes →