VP, AI Knowledge Engineering

IHG•Atlanta, GA

1d•Hybrid

About The Position

The Vice President, AI Knowledge Engineering will lead the design and delivery of the knowledge substrate on which every AI product in the enterprise depends — the ontologies that define our entities, the graph that connects them, the metadata that makes them discoverable, and the interfaces that make them safely accessible to agents. This is a build-and-transform mandate within the office of the SVP, AI & Engineering, with full ownership of the architecture and a multi-year horizon to get it right.

Requirements

Twelve or more years in knowledge engineering, enterprise data, or applied-AI platform leadership, with at least five years owning end-to-end delivery at scale.
Demonstrable experience designing and operating one or more of: enterprise ontologies, semantic layers, production knowledge graphs, or real-time data infrastructure — in a global or hyperscale operating environment.
Working fluency with the agentic-AI stack: model context interfaces, retrieval architectures, vector and graph stores, and the governance patterns that make them safe at enterprise scale.
Track record of leading large engineering and data organizations, including hiring, levelling, and developing senior technical talent.
Comfort operating with executive stakeholders — board, audit committee, regulators, owners, and franchise partners — on data, privacy, and AI risk.
The role owns five interconnected capabilities, delivered sequentially in year one and operated in parallel thereafter.

Nice To Haves

Public-company exposure: comfortable with disclosure discipline, segment reporting implications, and the cadence of investor communication.
Background in hospitality, travel, retail, or another consumer-scale industry where customer identity and real-time operational signals are core to competitive advantage.
Experience leading a transition from legacy batch and warehouse models toward streaming, graph, and agent-accessible architectures.
Direct experience designing or contributing to industry-level data standards, partnerships with hyperscalers, or external developer ecosystems.

Responsibilities

Define and govern the shared vocabulary of the enterprise — so every system, every model, and every agent shares one definition of guest, property, stay, and transaction.
Move from rows-and-tables to a relationship-first intelligence layer that links guest signals, property attributes, loyalty behavior, and operational events into a traversable graph that AI agents can reason over.
Tag the data estate with machine-readable ontologies, lineage, freshness indicators, and access classifications so AI systems can self-discover and trust enterprise data without human intermediation.
Stand up the Model Context Protocol layer and governed APIs through which internal and partnered AI agents query knowledge, trigger actions, and operate with full audit and policy control.
Replace batch dependencies with event-driven pipelines so the knowledge graph and every downstream AI consumer operate on current reality, not yesterday’s snapshot.