About Me

I'm a technology leader who builds enterprise-grade data and AI platforms end-to-end — real systems that ship, scale, and endure.

Over the last 20+ years, I've modernized legacy estates, reshaped operating models, and delivered products that hold up under real constraints: tight budgets, heavy regulation, messy data, and the inevitable gap between strategy slides and what actually gets built.

The last five years have been centered on AI. The last three have been almost entirely GenAI and agentic systems. I've watched AI move from "models in a lab" to decision-making embedded inside core workflows — and that's where the hard problems show up: reliability at scale, cost control, traceability, and blast-radius containment when something breaks. Just as important, it's where you need to explain outcomes clearly — to engineers, business leaders, and sometimes auditors.

My strength is connecting strategy to execution. I help leadership teams make better bets: what to build, what to skip, and how to sequence work so it produces measurable business outcomes — lower cost, faster throughput, stronger reliability, and governance that doesn't slow teams down.

I'm also comfortable in the weeds. I work on platform design, data architecture, cloud patterns, security and governance structures, and delivery practices that move ideas into production without grinding teams down. I frequently map business and technical capabilities, run maturity assessments, and build enterprise architecture functions from scratch — including pragmatic Architecture Review Boards that reduce risk and prevent every team from reinventing the wheel.

I stay hands-on. I review designs, debug systems, and code alongside teams when needed — because strategies only prove themselves in implementation details.

Outside work, I'm a relentless learner and builder. I read constantly and experiment constantly — code, cameras, audio systems, bikes, and whatever else grabs my curiosity. That curiosity shapes how I lead: building scalable operating models, developing strong technical leaders, and creating environments where teams deliver consistently excellent work.

🔬

Patent Application: AI Agent Observability

Filed patent application for a comprehensive AI agent observability framework that enables production-grade monitoring, tracing, and debugging of agentic AI systems.

The innovation addresses critical gaps in understanding and controlling autonomous AI agents in enterprise environments—covering decision traceability, cost attribution, quality measurement, and failure analysis across multi-step agentic workflows.

Core Expertise

Enterprise AI & GenAI

AI strategy, RAG architectures, agentic systems, evaluation strategies, production-grade reliability.

• RAG, agentic, and hybrid approaches selected by program needs—not one-size-fits-all
• Evaluation strategy (quality, latency, cost) and production gates that prevent "vibe-driven" deployment
• Reliability patterns: retries, idempotency, fallbacks, audit trails, and controlled blast radius
• Platform due diligence for strategic investment decisions—assessing technical fit, vendor lock-in, and TCO
• Multi-modal AI systems integrating text, vision, and structured data
• Prompt engineering, fine-tuning strategies, and model selection frameworks
• Security and compliance: data handling, PII protection, responsible AI standards
• Integration with existing enterprise systems and legacy modernization paths
• Cost optimization strategies balancing model performance with budget constraints
• Team enablement: training engineers on GenAI best practices and production patterns

Data & Governance

Data governance frameworks, MDM, medallion/lakehouse patterns, policy-driven access controls.

• Governance, stewardship, MDM, and quality programs that enable self-service and AI without chaos
• Medallion/lakehouse patterns, data mesh where it fits, avoiding architecture for architecture's sake
• Policy-driven access including AI-assisted policy generation from business-English requirements
• Decision rights, RACI, intake models, and operating rhythm that scales with the organization
• Data quality frameworks: profiling, lineage, observability, and remediation workflows
• Privacy and compliance: GDPR, CCPA, HIPAA alignment and automated compliance reporting
• Data catalog implementation and metadata management for discoverability
• Real-time and batch data processing pipelines with fault tolerance
• Data security: encryption, masking, tokenization, and access audit trails
• Cross-functional collaboration models with security, risk, legal, and business stakeholders

Cloud & Architecture

700+ application migrations, Kubernetes, service mesh, CI/CD, multi-team execution models.

• Cloud-first migration blueprints, refactoring paths, and "zero disruption" transition strategies
• Container orchestration: Kubernetes, service mesh (Istio, Linkerd), and sidecar patterns
• CI/CD pipelines with automated testing, policy guardrails, and deployment gates
• SLO-driven operations with observability, monitoring, and alerting strategies
• Multi-team execution models that scale delivery without breaking governance or standards
• Azure, AWS, GCP architecture patterns and cross-cloud portability strategies
• Infrastructure as Code (Terraform, Pulumi) and GitOps workflows
• API gateway patterns, event-driven architectures, and microservices decomposition
• Serverless architectures where appropriate, avoiding over-engineering
• FinOps: cost optimization, rightsizing, reserved instances, and budget governance
• Disaster recovery, business continuity, and multi-region architectures

Enterprise Architecture

Capability maps, investment alignment, rationalization, partnership-focused intake.

• Capability maps (business + technology) linking strategy to execution and investment decisions
• De-duplication of technical capabilities, platform consolidation, and cost reduction narratives
• Architecture intake and decisioning that accelerates delivery instead of blocking it
• Partnership model—collaborating with delivery teams, not running approval theater
• Application portfolio management: rationalization, modernization roadmaps, and sunset strategies
• Technology roadmapping aligned to business priorities and architectural principles
• Vendor evaluation, selection frameworks, and build vs. buy decision models
• Integration architecture patterns for complex, heterogeneous enterprise landscapes
• Security architecture: zero trust, identity management, threat modeling
• Acquisition due diligence: technical debt assessment, integration planning, and risk evaluation
• EA frameworks (TOGAF, Zachman) applied pragmatically—not as rigid dogma

Operating Model & Transformation

Sustainable operating models, intake/prioritization, cross-functional collaboration.

• Intake, prioritization, and delivery rhythm across portfolio, product, and platform teams
• Cross-functional collaboration models spanning security, risk, compliance, finance, and engineering
• Practical change management focused on adoption, training, and enablement—not slideware or theater
• Agile and DevOps transformation: team structures, ways of working, and cultural shifts
• Metrics and KPIs that drive accountability without creating perverse incentives
• Stakeholder engagement strategies for exec alignment and organizational buy-in
• Team topologies: platform teams, product teams, enabling teams, and their interactions
• Org design for technology functions balancing centralization and autonomy
• Training programs and enablement initiatives for upskilling technical teams
• Sustainable operating models that don't burn out teams or create technical debt

Experience

CohnReznick

Leader, EA, Data & AI

Current

Define and execute enterprise AI strategy and platform standards
Lead cross-functional teams of engineers and architects, develop next-level technical leaders
Built multiple large-scale autonomous AI applications across tax, accounting, and advisory practices
Platform due diligence: assessing technical fit, business alignment, and TCO for strategic AI investments
Established AI governance frameworks, responsible AI standards, and firm-wide platform guidelines
🔬 Patent pending: AI Agent Observability

Voya Financial

Sr. Director, Enterprise Architecture (Cloud, AI & Data)

Delivered first AI applications to production in early 2023—within months of Microsoft's AI product launch—achieving 70% cost savings for the marketing group
Built migration mechanisms supporting 700+ applications with repeatable patterns and risk controls
Led enterprise digital transformation across regulated environments
Instrumental in data strategy and enterprise platform modernization (integration and foundational capabilities)
Established capability maps (business + technology) to align investment decisions
Supported acquisition due diligence and integration planning

Deloitte

Manager, Cloud Strategy

Guided large-scale cloud transformation programs for Fortune 500 clients
Designed rapid re-platforming approaches and execution playbooks
Built pragmatic cloud governance that balanced speed with control

Virtusa

Director, Technology & Business Consulting

Led transformation programs across complex enterprises (financial services, utilities, education/media)
Built migration factory patterns and execution models for repeatable modernization
Delivered BPM and process modernization in regulated settings
Early AI and advanced analytics initiatives: modeling, forecasting, pattern recognition, and data strategy

Ventures

evidentAI

AI evaluation and observability platform focused on production-grade reliability. Built to help enterprises measure, monitor, and improve AI system quality with rigorous evaluation frameworks and real-time observability.

BetterDiagnostics

Healthcare AI platform leveraging advanced diagnostics and decision support. Applying enterprise AI patterns to clinical workflows with emphasis on accuracy, interpretability, and regulatory compliance.

AI Coder Bots Platform

Agentic coding platform that assists developers with code generation, refactoring, and technical debt management. Focused on enterprise developer productivity with governance and quality controls built in.

These ventures apply the same principles I bring to enterprise work: production realism, governance-first design, and systems that solve real problems rather than chase demos.