What I build
- Private LLM and RAG systems
- Decision-intelligence tools
- Workflow automation products
- Governed enterprise AI deployments
- Operational AI systems with aviation proof
About
I work at the intersection of AI engineering, enterprise operations, and production deployment.
My focus is building AI systems that survive contact with the real world — messy data, skeptical users, security constraints, internal politics, limited infrastructure, and pressure to show measurable value.
Much of my strongest work has been in aviation operations, where AI systems must support real workflows — not just produce impressive demos.
How I think
- Start with the decision and workflow
- Design for failure modes, review, and rollback
- Make sources, uncertainty, and limitations visible
- Measure adoption and operational usefulness
Technical strengths
- Full-stack product engineering
- RAG architecture and evaluation
- Backend systems, APIs, and data pipelines
- Dockerized deployment and reverse proxy operations
- Security, RBAC, auditability, and governance patterns
Enterprise constraints I understand
- Sensitive data boundaries
- On-prem and VPS deployment constraints
- Messy operational data
- Skeptical users and adoption risk
- Internal approval and governance workflows
Stack I trust
PythonTypeScriptNext.jsFastAPIReactPostgresRedisVector searchDockerNginxRBACSSOLLM evalWhisper/STTRAGDecision intelligence
What I am open to
Applied AI, enterprise AI architecture, LLM systems, aviation AI, and product-focused AI engineering opportunities where production judgment matters.