AI Architecture & Workflow Principal Engineer
I design AI-native systems and the developer workflows that build them.
I help teams turn AI from a coding shortcut into an engineering system: MCP infrastructure, agentic workflows, Claude Code skills, Spec-Driven Development, second-brain knowledge systems, and production-grade frontend architecture.
Featured case study
AI-driven defect recovery for a marketplace launch
I used Claude Code, Jira MCP, and cross-codebase analysis to cluster root causes, prioritize fixes, and help stabilize an insurance marketplace under a tight production deadline.
75
defects resolved
15
days to stabilize
< 30 min
root-cause clustering
AI Systems & Agent Architecture
I design how agents, tools, and data fit together — MCP servers, orchestration boundaries, and guardrails — so AI features are reliable and governable, not demos that break in production.
AI-Native Development Workflows
I turn AI into a system teams can depend on: Spec-Driven Development, reusable Claude Code skills and plugins, and a knowledge layer that makes AI-assisted delivery predictable.
Architect + Builder
Principal-level engineering that ships. I advise at the architecture level and write the production code — across fintech, health-tech, and enterprise teams, with US clients.
What I'm building
Shipped work behind the positioning — not slides.
MCP infrastructure
Production-shaped Model Context Protocol servers you can run and deploy.
Learn more →Agentic workflows
An autonomous agent trading real money — and what the architecture, not the P&L, taught me.
Learn more →Spec-Driven Development
Specs as the source of truth AI builds against — predictable delivery, not improvisation.
Second Brain
An Obsidian + Claude Code knowledge workflow that compounds over time.