Enterprise AI Platform

Enterprise-scale AI platform architecture for Disney Experiences

The Walt Disney Company � Associate Architect / Senior GenAI Analyst � Dec 2024 � Present

Context

Disney Experiences operates at an intersection of world-class storytelling, large-scale operations, and stringent data governance requirements. Translating AI capabilities into production-grade enterprise systems in this environment requires more than technical knowledge � it requires the ability to navigate governance, compliance, legacy architecture, and cross-functional stakeholder expectations simultaneously.

Architecture Work

AI Platform Vision & Roadmaps

Defined and communicated an enterprise AI platform architecture vision across application and integration domains. Developed multi-year adoption roadmaps from large use-case inventories, prioritizing by business value, risk profile, and regulatory readiness. Established architecture standards, patterns, and review frameworks for agentic and retrieval workloads � enabling consistent, reusable, and auditable AI capability delivery across brownfield and greenfield systems.

Authored architecture decision records (ADRs), integration specifications, and developer guides to formalize governance artifacts and enable structured stakeholder review.

Multi-Model AI Gateway

Architected and delivered an enterprise-scale AI gateway spanning multiple frontier model providers (including Claude, GPT-4o, Gemini, and Nova-class models). The gateway provides virtual keys, spend controls, guardrails, PII handling, and observability callbacks � production-deployed with measurable cost and compliance outcomes.

Governed RAG & Semantic Search

Led end-to-end solution architecture for enterprise semantic and document search integrated with grounded retrieval-augmented generation (RAG). The system delivers traceable, internally governed answers, replacing ad-hoc search with auditable, source-attributed responses aligned to enterprise data governance requirements.

AI Workspace

Designed and implemented an internal AI workspace with SSO/RBAC and model-level access controls. Tool surfaces are registered and governed with least-privilege and auditable patterns, enabling controlled access to AI capabilities across roles and teams.

Orchestration & IaC

Built and governed no-code/low-code orchestration (n8n) for ingestion, transformation, and retrieval pipeline handoffs. Provisioned platform infrastructure via GitLab CI and Terraform � repeatable, security-scanned infrastructure-as-code supporting a DevSecOps delivery model.

Spec-Driven Development

Implemented spec-driven development (SDD) at the platform level: specifications, contracts, and architecture review gates before build; delivered production agentic workloads with full lifecycle ownership.

Approach

Architecture in an enterprise AI context is fundamentally a governance problem as much as a technical one. Every capability � a RAG system, an agentic workflow, a model API call � must be designed with auditability, rollback, least-privilege, and compliance traceability as first-class requirements, not afterthoughts.

The work at Disney reflects that approach: not bolting AI onto existing systems, but designing AI-native platform infrastructure that is governed from the ground up and can support a durable, multi-year capability roadmap.

Non-confidential summary. Specific internal system names, model configurations, and proprietary architectural details are not disclosed.