Oracle
AI strategy for enterprise construction project management.
Five years at Oracle building AI strategy inside a SAFe-agile enterprise platform. The product managed construction projects worth billions of dollars.
This is the work that required understanding retrieval-first AI before it had a name.
Enterprise AI strategy, five years in
I built the AI roadmap for Unifier inside Oracle's construction software suite. That meant navigating SAFe-agile, enterprise procurement cycles, and a codebase that predated modern AI primitives.
- Retrieval-first AI roadmap (5yr horizon)
- SAFe-agile methodology at scale
- Stakeholder alignment across enterprise PMs and engineers
AI in the construction vertical
Construction software manages projects worth billions, runs on 20-year-old data models, and is ripe for AI-native redesign. I spent five years understanding what that actually requires.
- Enterprise AI adoption patterns
- Construction tech market dynamics
- Retrieval-first architecture in legacy platforms
Retrieval-first in an enterprise codebase
Building a retrieval-first AI roadmap inside Oracle Unifier meant working around a relational data model that was never designed for vector retrieval. The constraints were instructive.
- Retrieval-first roadmap in a SQL-native platform
- AI architecture inside SAFe agile delivery
- Tradeoffs between model capability and enterprise compliance
Five years inside enterprise construction software
I helped build AI strategy for a platform that manages real construction projects at scale. This is that story.
Tell me what you are looking for
Oracle context spans AI strategy, enterprise product, and construction tech. Let me know your angle.