Jason Curry

Hands-on Delivery • Data Center Programs • Platform Modernization • Distributed Systems

Hands-on Technical Program Manager with 16+ years driving end-to-end execution of infrastructure, data center, and platform modernization programs across enterprise and telecom-scale environments. Known for operating directly in the delivery path—owning execution tracking, dependency resolution, release coordination, and production readiness activities across complex distributed systems.

Extensive experience leading data center onboarding, multi-region infrastructure integration, API/middleware ecosystem alignment, and cloud modernization initiatives with direct accountability for program delivery outcomes. Regularly embedded with engineering and infrastructure teams to drive execution through ambiguity, production constraints, and cross-system dependencies.

Core execution domains:

• Data Center Onboarding & Infrastructure Cutover Execution
• Platform & Distributed Systems Delivery Coordination
• Release Train & Cross-Team Engineering Execution
• Cloud Migration & Hybrid Infrastructure Modernization
• Operational Readiness, Observability, and Incident Coordination
• AI-assisted delivery tracking and impact analysis workflows

Data Center & Infrastructure Delivery

• Drive end-to-end data center onboarding execution including network, security, and platform integration
• Coordinate firewall rules, segmentation, and secure zone implementation with engineering/security teams
• Own dependency mapping across infrastructure, middleware, and application layers
• Execute multi-region rollout sequencing and cutover planning
• Validate operational readiness for production integration

Platform & Distributed Systems Execution

• Drive execution across middleware and API ecosystems in distributed environments
• Coordinate microservices and platform integration work across multiple engineering teams
• Resolve cross-team blocking issues impacting system dependencies and releases
• Manage system-level integration risks during modernization efforts

Program Delivery & Release Execution

• Run large-scale release train execution across distributed engineering organizations
• Own program-level delivery tracking, risk escalation, and dependency resolution
• Drive CI/CD release coordination across multiple concurrent engineering streams
• Lead production readiness reviews and go-live execution checkpoints
• Ensure delivery alignment across infrastructure, security, and application teams

Operations, Reliability & Production Support

• Drive SLO/SLI tracking alignment with engineering and operations teams
• Use observability tools (Splunk and others) to monitor production health and execution risk
• Coordinate incident response, change management, and production stabilization efforts
• Ensure operational readiness before and after production cutovers

Cloud & Modernization Execution

• Execute cloud migration and hybrid infrastructure modernization programs
• Coordinate sequencing of workloads across legacy and cloud environments
• Manage migration dependency chains and production transition plans
• Support infrastructure modernization roadmap execution

AI & Automation (Delivery Enablement)

• Apply AI-driven analysis for program impact assessment and dependency tracking
• Improve visibility into execution risk, migration readiness, and delivery throughput
• Support automation of program reporting and operational intelligence workflows

📊 Case Study: Capacity Planning & FinOps Intelligence Platform (DOMO)
Mastercard | Engineering Infrastructure & Platform Systems

🧭 Overview
Led the design and delivery of a unified Capacity Planning & FinOps intelligence platform built in DOMO, integrating infrastructure telemetry, workload utilization, and cost attribution across Mastercard’s engineering hardware and distributed infrastructure environment.

This initiative enabled a shift from reactive reporting models → predictive, telemetry-driven capacity planning, improving visibility into compute consumption, workload demand, and infrastructure cost efficiency at enterprise scale.

🚨 Problem Statement
Engineering and infrastructure teams lacked a consistent, scalable system to:

• Track real-time capacity utilization across distributed compute environments
• Correlate workload demand with infrastructure consumption patterns
• Attribute cost drivers across shared engineering infrastructure
• Forecast future capacity requirements with reliability
• Align engineering consumption with FinOps governance models

Key limitations included:
• Fragmented reporting across spreadsheets and static dashboards
• No unified telemetry-driven capacity model
• Limited workload-level cost attribution
• Reactive scaling decisions instead of predictive planning

🎯 Objectives
• Establish a centralized Capacity + FinOps intelligence system in DOMO
• Integrate distributed infrastructure telemetry into a unified data model
• Enable workload-level visibility into utilization and cost drivers
• Improve forecasting accuracy for capacity planning cycles
• Support FinOps-driven optimization and cost governance initiatives

🏗️ Solution Architecture
• Built a telemetry ingestion layer aggregating infrastructure signals (compute, storage, network, workload metrics)
• Designed a unified capacity modeling framework spanning allocated vs. actual utilization
• Implemented workload-level cost attribution mapping compute usage to financial drivers
• Developed DOMO-based analytics layer for real-time dashboards and executive reporting
• Created forecasting models using historical utilization trends and workload growth patterns
• Enabled drill-down visibility from enterprise → cluster → workload → service level

📈 Key Outcomes
• Improved visibility into enterprise-wide infrastructure utilization
• Transitioned capacity planning from reactive → predictive model
• Identified underutilized compute clusters and optimization opportunities
• Strengthened FinOps alignment between engineering consumption and cost governance
• Enabled ~30% infrastructure cost optimization through improved attribution and visibility
• Increased forecasting accuracy for capacity planning cycles

🧠 Key Technical Contributions
• Designed telemetry-driven capacity planning architecture at enterprise scale
• Built workload-to-cost attribution framework enabling FinOps alignment
• Developed utilization-based allocation models for shared infrastructure systems
• Embedded FinOps principles into engineering analytics and reporting layer
• Positioned DOMO as an operational decision intelligence platform (not just reporting)

🔑 Core Keywords
Cloud Infrastructure • Capacity Planning • FinOps • Cost Attribution • DOMO • Telemetry • Observability • Distributed Systems • Kubernetes • Microservices • SRE • Cloud Cost Optimization • Infrastructure Engineering • Predictive Analytics • Engineering Platforms

💡 Key Insight
Capacity planning becomes strategically powerful when driven by real-time telemetry and workload-level attribution—transforming infrastructure from a static cost center into a continuously optimized, data-driven system.

📌 Case Study: AI-Driven Impact Analysis Platform (Southeast Toyota Finance) 🧭 Overview 🚨 Context & Problem 🎯 Core Problem 👤 Role & Scope 🏗️ Approach Knowledge Ingestion & Context Layer Dependency Mapping (Knowledge Graph) LLM + RAG Architecture Workflow Integration 📈 Results & Impact 🧠 Key Insight 🔑 Strategic Takeaways

Led design and delivery of an AI-powered impact analysis platform that automated downstream system and business impact detection across enterprise applications, data pipelines, and compliance workflows.
Replaced manual, SME-dependent analysis with a scalable, AI-assisted decision system leveraging LLMs, retrieval-augmented generation (RAG), and system dependency mapping.

Enterprise change events (pricing, policy, compliance updates) required deep analysis across interconnected systems.

Key challenges included:
• Manual impact analysis reliant on tribal knowledge
• Lack of system-to-system dependency visibility
• High risk of missed downstream impacts
• Slow decision cycles (days to weeks)
• Inconsistent documentation across APIs, schemas, and workflows
• No centralized mechanism to validate change risk before deployment

Design a scalable, intelligent system capable of identifying downstream impact across systems while improving:
• Speed of analysis
• Accuracy and completeness
• Repeatability and auditability

Principal Technical Product Owner / AI Platform Lead

• Owned product strategy and architecture for AI-driven impact analysis
• Defined system ontology and dependency mapping model
• Led design of LLM + RAG architecture
• Partnered with engineering, data, and compliance teams
• Integrated solution into CI/CD and change management workflows

• Aggregated enterprise artifacts:
– API specifications (OpenAPI)
– Database schemas and lineage
– Business rules and policy documents
• Built vectorized knowledge base using embeddings

Outcome:
Created a unified, queryable enterprise knowledge layer

• Modeled relationships across:
– Applications
– APIs
– Data pipelines
– Business processes
• Enabled graph traversal for impact reasoning

Outcome:
Shifted from document search → system-level reasoning

• Designed prompt orchestration layer
• Injected contextual data dynamically from:
– Vector search
– Graph traversal results
• Constrained outputs into structured formats

Outcome:
Produced deterministic, explainable impact analysis results

• Integrated into CI/CD pipelines for pre-deployment checks
• Embedded into change management workflows
• Enabled human-in-the-loop validation

Outcome:
Moved impact analysis from reactive → proactive

• Reduced impact analysis time: days → minutes
• Increased confidence in change planning and approvals
• Improved detection of downstream dependencies
• Reduced production risk from incomplete analysis
• Enabled scalable, repeatable decision-making

Improvements across:
• Speed
• Accuracy
• Risk management
• Operational efficiency

The core problem was not lack of intelligence—it was lack of structured context.

By combining:
• retrieval (RAG)
• relationship modeling (graph)
• controlled inference (prompt orchestration)

AI became reliable enough for enterprise decision workflows.

• LLMs without context are not enterprise-ready
• Knowledge graphs significantly enhance AI reasoning capability
• Embedding AI into workflows is more valuable than standalone tools
• Deterministic output design is critical for trust and adoption

Led the design and implementation of an Enterprise AI Governance and Orchestrator Platform supporting a large-scale AS/400-to-cloud modernization program.

The initiative began as a real-time AI Decision Intelligence platform intended to reduce operational decision latency and evolved into a governed AI operating model capable of safely scaling generative AI and agent-based workflows across regulated modernization initiatives.

The platform combined real-time decision intelligence, policy-driven governance, AI agent orchestration, and downstream impact measurement into a closed-loop modernization intelligence system.

Modernization planning and migration decision analysis required 4–8 hours per request due to fragmented data sources, manual analysis workflows, and limited observability across legacy and cloud environments.

As AI adoption accelerated, additional risks emerged:

• Inconsistent governance across AI workflows
• Lack of standardized prompt controls
• Unmanaged agent execution
• Limited visibility into downstream client impact
• Regulatory and auditability requirements during migration

A unified AI operating model was required to safely scale AI-assisted modernization activities.

Designed a real-time decision intelligence platform integrating:

• Legacy AS/400 operational systems
• Modern cloud services
• Event-driven data pipelines
• Workflow automation services
• Enterprise observability systems

The platform automated modernization decision analysis and provided near real-time insights supporting migration planning and execution.

Expanded the architecture into a governed enterprise AI ecosystem.

The Orchestrator served as the central control plane responsible for:

• Intent classification
• Governance evaluation
• Policy enforcement
• Agent routing
• Workflow coordination
• Response validation
• Audit logging

The Governance Framework enforced:

• Policy-as-Code controls
• Data classification requirements
• RBAC and ABAC authorization
• Prompt guardrails
• RAG grounding controls
• Output validation
• Compliance monitoring
• End-to-end traceability

Every AI interaction passed through governance evaluation before execution.

Implemented a formal impact measurement framework to evaluate modernization outcomes beyond technical migration metrics.

Measured dimensions included:

• Decision cycle reduction
• Planning acceleration
• Reduction in manual analysis

• Service stability
• Business continuity
• Responsiveness improvements
• Reduction in migration disruption

• User adoption rates
• Repeat usage patterns
• Operational integration

• Audit readiness
• Policy adherence
• Data lineage coverage
• Decision traceability

DCIA outputs continuously informed migration prioritization, orchestration logic, governance policy tuning, and modernization sequencing.

• 90% reduction in decision latency
• Decision cycles reduced from 4–8 hours to minutes and seconds
• Governed AI deployment across modernization workflows
• Centralized orchestration of enterprise AI agents
• Improved observability and audit readiness
• Increased executive confidence in modernization initiatives
• Closed-loop operating model connecting AI decision intelligence, governance enforcement, orchestration, and downstream client impact measurement

This platform established the foundation for scalable enterprise AI adoption while supporting a high-risk legacy-to-cloud transformation program.