Open to full-time VP of Technology, CTO, CIO, and Director of IT roles — remote

Called in when a company is scaling, acquiring, modernizing, or recovering

I turn fragmented, post-acquisition technology estates into one operating platform.

Growth breaks how a company runs. I rebuild it without pausing the business.

Eighteen years leading enterprise IT infrastructure, security, and transformation for a 2,000-person, nine-country organization — through eleven acquisitions, a ransomware event, and a full colocation exit. Now building AI agents that run in production, not in slide decks.

Eighteen years absorbing what growth does to an operating model — eleven acquisitions, a ransomware event, a full data-center exit, and a global workforce moved remote in under fourteen days. Every one of them executed while client delivery and revenue continued. Now building AI agents that run in production, not in slide decks.

Review the track record joel@reyes.global

See the decisions How I would start

Joel A. Reyes

Enterprise ITSecurityAI Operations

Now: Fractional VP of IT, Prinox
Based: New Jersey, US — global teams
Called in: Scaling, acquiring, modernizing, or recovering
Engagement: Full-time only — remote, up to 20% travel

11×

acquisitions integrated while scaling from 200 to 2,000+ employees

$2.7M

annual colocation spend eliminated — ~70% net infrastructure cost reduction

<36h

ransomware recovery as incident commander, with zero data loss

160+

initiatives delivered cross-functionally with zero operational downtime

11×

acquisitions absorbed while the business kept delivering — 200 to 2,000+ employees

$2.7M

taken out of the annual operating base, at ~70% net infrastructure cost reduction

<36h

from ransomware event to recovered, zero data loss, as incident commander

operational downtime across 160+ cross-functional initiatives delivered

Point of view

Most failed AI transformations are not model problems. They are execution problems.

AI does not fix a broken process. It scales it. When a workflow has no owner, no clean data, and no definition of done, automation just produces the same defect faster and with less accountability. The pilot demos well, the rollout stalls, and the board concludes the technology was overhyped.

The unglamorous work first

Map the workflow. Name the owner. Fix the data. Then automate the constraint.

That is operations work, not data-science work — which is why AI programs belong to the technology executive accountable for how the company actually runs.

I approach it the way I approached eleven acquisition integrations: find the constraint, decide in public, instrument the result, and tie every automation to a number a CFO recognizes.

Workflow before model

Define the process, the owner, and the exception path before choosing a tool.

Governance is the enabler, not the brake

Row-level security, secret redaction, rate limits, and audit trails are what let an agent touch production data at all.

Measured in operating terms

Cycle time, throughput, cost per transaction — not tokens, not prompts, not pilots.

Track record

Four decisions that defined the environment.

Four decisions and what they returned.

Each one is a real operating call under constraint — the situation as inherited, the decision made, and the outcome it produced.

Each one is a real call made under constraint, with money or continuity on the line — what was inherited, the decision taken, and what it cost or saved.

Exiting the data center without betting the business

Infrastructure modernization

Situation

Three legacy colocation providers carrying $2.7M in annual spend, supporting client delivery that could not pause for a migration window.

Decision

Full colocation exit with roughly 30% of the freed spend deliberately redirected into Azure and AWS — a rebuild of the platform, not a lift-and-shift of the same problems.

Result

$2.7M in annual colocation spend eliminated at a stated ~70% net reduction in infrastructure operating cost.

Incident command during a ransomware event

Resilience & security

Situation

A ransomware event against a global research organization where recovery objectives had historically been measured in days.

Decision

Took incident command, then rebuilt the recovery program around tested restore paths, identity containment, and a hardened backup posture rather than a documentation exercise.

Result

Recovery in under 36 hours with zero data loss, and RTO reduced from days to under 36 hours as a standing capability.

Eleven acquisitions, one operating platform

Making the twelfth acquisition easier than the first

M&A IT integration

M&A value capture

Situation

Growth from 200 to 2,000+ employees across 9 countries, 4 continents, and 40+ locations — each acquisition arriving with its own identity, network, and vendor stack.

Growth from 200 to 2,000+ employees across 9 countries, 4 continents, and 40+ locations — every deal arriving with its own identity, network, and vendor stack, and its own integration bill.

Decision

Standardized integration playbooks covering identity consolidation, platform migration, office buildouts, and three enterprise-wide rebrands, executed by a 42-person global team across the US, EMEA, APAC, and India.

Replaced one-off integrations with a standardized playbook — identity consolidation, platform migration, office buildouts, three enterprise-wide rebrands — so each successive deal absorbed faster than the one before. Run by a 42-person team across the US, EMEA, APAC, and India.

Result

160+ cross-functional initiatives delivered with zero operational downtime, and $750K+ in annual vendor savings against a $10M+ technology budget.

160+ cross-functional initiatives delivered with zero operational downtime, and $750K+ in duplicate vendor spend converted into recurring annual savings against a $10M+ technology budget.

A global workforce moved remote in under two weeks

Business continuity

Situation

A distributed, facility-dependent research operation that had to keep delivering for clients when its offices closed.

Decision

Prioritized identity, endpoint, and secure remote access over a full re-architecture — sequencing for continuity first, cleanup second.

Result

Fully remote in under 14 days with no interruption to client delivery or revenue.

Building now — Prinox

A twelve-agent research platform, in production.

Prinox is a self-funded, AI-enabled market research automation platform. It exists because the argument above needed proof: an executive who claims AI belongs to operations should be able to ship an agent system that survives real data, real clients, and real governance requirements.

Stage 01

Intake

Client objectives captured and structured into a research brief.

Stage 02

Research design

Methodology, sample framing, and instrument construction.

Stage 03

Synthesis

Multi-agent analysis with source traceability across findings.

Stage 04

Reporting

Executive and operational briefings generated for delivery.

Production controls

Row-level security Secret redaction Rate limiting Signed API gateway Scoped module keys Idempotency Outbox events Shadow-mode consumers Nightly backups Retention controls

Stack

TypeScript Vercel Supabase / PostgreSQL n8n Anthropic Claude OpenAI GitHub Actions Python

A live example: a daily CFO workflow that pulls platform, usage, and financial data, calculates burn rate, reads billing inboxes, applies two-tier invoice alerts, and books a same-day review hold when something needs a human.

Ask for a walkthrough

First ninety days

What I would do before asking you for budget.

No reorganization in week one. Three phases, and a number you can hold me to at the end of them.

Days 1–30

Find the constraint

Interview the people who touch the work, not only the leaders who describe it. Map where cycle time actually goes, what the technology base genuinely costs, and which risks are one incident away from becoming your problem. Output is a ranked list of constraints with an owner named against each.

Days 31–60

Stop the measurable bleeding

Fix what needs no committee: duplicate vendor spend, unowned access, untested restores, the manual step everyone quietly works around. Early credibility comes from removing pain the organization can feel, not from presenting a roadmap.

Days 61–90

Commit to a number

One or two initiatives, sequenced, each carrying a target a CFO recognizes — cost per transaction, cycle time, recovery objective, integration days-to-value. Instrumented before launch, so progress is observable without anyone having to ask me.

Operating scope

The size and shape of environment I have run.

Not aspiration — the actual perimeter of organization, footprint, team, budget, and compliance obligation carried in role.

Useful mainly as a floor. If your environment is smaller, none of this breaks; if it is larger, this is the closest comparable I can offer you.

Organization: 2,000+ employeesgrown from 200 through acquisition
Footprint: 9 countries, 40+ locationsacross 4 continents
Team: 42 global staffHelp desk, security, data, network, cloud — 4 direct reports
Budget: $10M+ annualtechnology budget and vendor portfolio
Compliance: ISO 27001certification program owned for roughly 8 years
U.S. Department of Defense: Joint Task Force Guantanamo BayClassified, continuously inspected network — 24/7 court-schedule uptime, 11,000+ users at peak
Continuity: Under 14 daysto move a global workforce fully remote

Leadership & strategy

Executive leadership IT strategic planning Technology roadmapping Digital transformation Mergers & acquisitions Enterprise architecture Program management Vendor management Budget & financial operations

Cloud, infrastructure & security

Azure AWS Microsoft 365 Entra ID / hybrid identity VMware & Hyper-V SD-WAN / MPLS Zero Trust architecture IAM, MFA, privileged access Cybersecurity governance IT risk management SOC 2 & GDPR DR / business continuity NIST & DISA STIG ITIL / ITSM

Automation & engineering

Multi-agent orchestration n8n Anthropic Claude OpenAI Microsoft Security Copilot PowerShell Terraform Intune CI/CD & GitHub Actions TypeScript SQL / PostgreSQL Python

Experience

Eighteen years, one continuous operating thread.

Built from the network layer up — which is why the strategy holds when it meets the infrastructure.

Jan 2026 — Present

Fractional VP of Information Technology

Prinox

Founded and built a self-funded AI research automation platform — twelve agents spanning intake, research design, synthesis, and reporting, with production security and governance controls.

Oct 2020 — Dec 2025

Senior Director, Information Technology Infrastructure

Sago

Owned global infrastructure, security, and operations through eleven acquisitions, a colocation exit, a ransomware recovery, and three enterprise rebrands.

May 2014 — Oct 2020

Senior Network Administrator

Sago (formerly Schlesinger Group)

Standardized network, identity, and platform onboarding across 40+ global locations — the operational foundation later scaled into the repeatable M&A integration playbook.

Jan 2011 — Oct 2013

Systems Engineer

Joint Task Force Guantanamo Bay — U.S. Department of Defense

Compliance point of contact for a classified, continuously inspected network supporting 11,000+ users at peak, under 24/7 court-schedule uptime obligations. Translated technical risk into terms auditors and command leadership could act on.

Sep 2009 — Jun 2011

Systems & Network Administrator

DSCI, a TelePacific company (now TPx Communications)

Led a data-center migration and architected the full network wiring design — infrastructure built to let the business scale, not merely to keep it running.

Education & credentials — B.S., Information Technology, Western Governors University. Technical Diploma, Information Technology, Anthem Institute. CompTIA Security+, Network+, and Server+; CNIP; Jira Service Management Fundamentals. Currently working toward the CISSP.

Education & credentials

B.S., Information Technology — Western Governors University. Technical Diploma, Information Technology — Anthem Institute.

CompTIA Security+ CompTIA Network+ CompTIA Server+ CNIP Jira Service Management Fundamentals CISSP — in progress

For executive search & hiring teams

For CEOs, boards, and sponsors

Let's talk about the environment you need fixed.

Tell me what growth broke.

joel@reyes.global Connect on LinkedIn

I respond to every serious inquiry personally. If you are evaluating me for a VP, CTO, CIO, or Director role, tell me the situation you are hiring against — integration backlog, cost pressure, security exposure, an AI mandate with no operating model — and I will tell you honestly whether I am the right operator for it.

I respond to every serious inquiry personally. Bring me the situation rather than the job description — an integration backlog, a cost base growing faster than revenue, a security exposure you cannot price, an AI mandate with no operating model. I will tell you honestly whether I am the right operator for it. I am looking for one full-time seat, not a portfolio of engagements.

Target roles: VP of Information Technology · Chief Technology Officer (CTO) · Chief Information Officer (CIO) · Director of IT Operations
Engagement: Full-time only — not seeking fractional or advisory work
Work model: Remote, US-based, open to travel up to 20% — experienced running US, EMEA, APAC, and India teams

Where I am useful: Scaling, acquiring, modernizing, or recovering — where the operating model has not caught up with the growth
The seat: A full-time executive seat — VP of Information Technology, Chief Technology Officer (CTO), Chief Information Officer (CIO), or Director of IT Operations. Not fractional, not advisory.
Reach: Remote, US-based, open to travel up to 20% — experienced running US, EMEA, APAC, and India teams