X-energy / advanced nuclear
Enterprise AI Systems
Enterprise software and agentic workflows built where engineering rigor, security boundaries, and operational reliability are non-negotiable.
OUTCOME / 01
Turned domain expertise into a software role by repeatedly shipping useful systems from inside the engineering workflow.
deployment context
ownership
domain
The work
I build enterprise AI and engineering systems at X-energy. The work spans agentic workflows, platform reliability, developer systems, analytics, and domain-specific applications used to make engineering work faster and more traceable.
How I approach it
Advanced-nuclear software has a useful constraint: a convincing demo is not enough. The system needs explicit permissions, inspectable state, failure handling, tests, and a path for a human to understand what happened.
- Start from the real engineering or business decision, not the model.
- Keep identity, authorization, and tenant boundaries explicit.
- Stream useful progress while preserving cancellation and recovery.
- Treat evaluation, observability, and CI as product features.
Public boundary
The underlying systems are internal. This page intentionally describes the engineering method and scope without publishing confidential data, topology, prompts, or screenshots.