Technology & AI Program Consultant / Discovery to Production
I lead complex technology and AI programs end to end.
I'm Alex Henry, a technology and AI consultant in Atlanta. For 20+ years I led technology programs at Delta Air Lines, Intuit Mailchimp, and in the Marine Corps, taking work nobody had scoped yet all the way into production. I do that for clients now. I also run companies on software I write myself, which is a harder test than any reference call.
Why me
Most people advising on this have either run large programs or run a business. I have done both, and I still do.
I spent 20+ years accountable for delivery at real scale. Enterprise notification infrastructure at Delta Air Lines. AI program delivery across a 34,000+ customer base at Intuit Mailchimp. IT strategy and modernization in the Marine Corps, where downtime was not an option. Then I built companies of my own, and I write the software they run on.
What I do
Three ways in. Same discipline underneath.
Some organizations know exactly what they need built and have nobody free to own it. Others know AI should be doing more but cannot tell where. Most have a tangle of systems that will fight both. I take on all three, and in practice they usually turn out to be the same job.
01 of 03 / AI Strategy, Enablement & Systems
AI that gets understood, adopted, and used.
Most organizations do not have an AI problem. They have an adoption problem. Someone bought the tools, ran a pilot, and the team quietly went back to the old way of working. So I start with a straight conversation about what AI can and cannot do in your operation, then build the few systems that earn your team's trust. Understanding first. Systems second. I led enterprise AI delivery across 12 engineering, product, and data teams at Intuit Mailchimp, and I still live with the results of my own builds every day.
AI training for your team
Working sessions that get people past both the hype and the fear. They leave knowing what to hand to AI, what to keep doing themselves, and how to tell the difference.
Where AI actually pays
I map your operation and find the handful of places AI earns its keep. Just as important, I tell you where it will not, before you spend money finding out.
Custom assistants and automation
Assistants built on your data, your voice, and the way your team actually works. Not a generic chatbot bolted onto the side of the business.
Systems integration
I connect the software you already pay for, so information gets entered once and the steps after it happen on their own.
The third column is the one that saves you money. Most AI budgets die on work that should have been left alone.
Illustrative of the method, not client data.
Delivered at Intuit Mailchimp and in my own companies your numbers scoped in the first call.
02 of 03 / Technology Program Leadership
For the program nobody on your team can fully own.
Some work is too cross-functional for one team, too ambiguous for a project plan, and too important to keep slipping. That is the work I have done for 20+ years. I come in, get my arms around what is really going on, and carry it from a vague ask to something running in production with your name on it.
Discovery and architecture
I map what you are actually running before I propose anything. Every system, every integration, every workaround somebody built because the real tool did not work. Then I design the target state and the sequence to get there.
Delivery leadership
I run the program across engineering, product, data, and whichever vendors are involved. Dependencies stay visible, risk gets named early, and the people doing the work know what good looks like this week.
Vendors and build-versus-buy
I have managed a portfolio of 68 integration vendors, so I know where custom architecture beats off-the-shelf and where it is an expensive mistake. You get the recommendation and the reasoning, not a preferred partner.
Rollout and handover
The program is not done when it ships. It is done when your team owns it, the runbook is real, and nothing breaks quietly two quarters later. I stay until that is true, then I leave.
Where I learned it
Risk, named early
Legacy export format blocks cutover. Owner assigned, decision due Friday.
Representative of the format, not client data.
Delivered at Delta Air Lines your program gets sized in the first call.
The proving ground
Everything I sell, I run first.
I did not stop building when I left the enterprise. I founded companies and I write the software they run on, which means every recommendation I make has already been tested somewhere the failure lands on me. 170+ automations running today, roughly $52K a year of manual work automated out, and about 20 hours a week handed back.
I mention this for one reason. Most people at my level of the org chart stopped touching production a decade ago. I never did.
The proof
I would rather show you numbers than make you promises.
Two decades of it is in the enterprise record below. The rest is work I did for people who agreed to put their name on the result.
Teams I have built for
Alex did not hand us a strategy deck. He built the thing and it was live in weeks. Our team stopped losing three hours a day to manual data entry.
None of it was theory. He set it up, showed my team how to use it, and stayed with us until it stuck.
The upsell messaging paid for the whole engagement before we even finished. Guests get quicker answers and we stopped leaving money on the table.
The approach
Every program runs the same four phases. The order is the whole method.
I do not show up with a ready-made framework and try to make you fit it. What is fixed is the sequence, because skipping discovery to get to building faster is the single most reliable way to spend a year on the wrong thing. Everything else bends to your organization.
Discovery
Before I recommend anything, I map what you are actually running. Every system, every vendor contract, every integration, and every workaround somebody built because the real tool did not work. Most programs go wrong right here, because they start from the org chart's version of the truth instead of what the business actually does day to day.
Design and sequencing
Then I put a plan in front of you, sized to the team you actually have. What gets built, what gets bought, what gets retired, what order it all happens in, where the risk sits, and what it should pay back. You approve that plan before anyone writes a line of code.
Delivery
I run the build across engineering, product, data, and vendors. Dependencies stay visible and problems get named while they are still cheap. This is the part that separates a program that ships from a program that gets quietly rescoped at the end of the quarter.
Handover
The engagement ends when your team owns the thing. Real documentation, real runbooks, and the people who have to live with it knowing why it was built this way. I would rather hand you the keys than sell you a dependency.
Questions
The things people ask before we talk.
What does Alex Henry do?
I am a technology and AI consultant. I lead complex programs from discovery through production, which covers three things: AI strategy and team enablement, end-to-end program leadership for work that is too cross-functional for any one team to own, and the systems integration and modernization that usually has to happen underneath both. I spent 20+ years doing this at Delta Air Lines, Intuit Mailchimp, and in the U.S. Marine Corps before I started doing it for clients.
What is your background?
I was Principal Technical Program Manager at Delta Air Lines, where I owned the enterprise customer-notification infrastructure and delivered $12M in measured business impact. Before that I was Senior IT and Software Engineering Manager at Delta, leading a 172-person engineering organization across 3 countries with $384M in awarded contracts under my ownership. At Intuit Mailchimp I was Staff Technical Program Manager for enterprise AI delivery, reaching a 34,000+ customer base. Before all of it, 16 years in the U.S. Marine Corps leading IT strategy and modernization. I hold a PMP, a Lean Six Sigma Black Belt, an AWS certification, and an MBV from USC Marshall.
What kinds of programs do you take on?
Usually the ones that have stalled. That tends to mean an AI project stuck after the pilot, a cloud migration or legacy modernization nobody has been able to sequence, or an integration program spanning more teams and vendors than one manager can hold. If your engineers are good but no single person is accountable for the whole outcome, that is the gap I fill.
What makes your approach different?
Two things. I start with discovery rather than tools, because most AI programs fail on adoption instead of technology. Someone buys the software, runs a pilot, and the team quietly goes back to the old way. And I still build. I run companies on software I write myself, so I am not recommending architecture I have never had to maintain at 2am.
Do you only work in hospitality?
No. The core work is technology and AI program delivery, and it applies to any organization with a program bigger than the people available to run it. I do own and operate businesses in hospitality and vehicle rental, and I write the software they run on, but those are where I prove the method. They are not the market I serve.
What is Guestally?
Guestally is an AI guest-communication and upsell platform I took from concept to production, and it sells today. I point to it because it is the cleanest answer to whether a program leader can still ship. I did the architecture, wrote the software, and put it in front of paying customers.
How much does it cost and how long does it take?
It starts with a short call, on me, to figure out whether I can actually help. I do not put anyone on a retainer until the scope is clear and the payback makes sense. Timeline depends entirely on the program, and I would rather size it honestly after discovery than quote you a number I would have to walk back.
Where are you based and who do you work with?
I am based in Douglasville, Georgia, just outside Atlanta, and I work remotely with clients across the US. I take on organizations that have a technology or AI program that matters and needs one person to own it end to end.
Let's talk
The fastest way to know if I can help is a short call.
Tell me what you are trying to solve and I will tell you honestly whether I am the right person for it. I only take on a handful of clients at a time, and I read every message myself.
Where I share what I am learning about AI, automation, and running a hospitality business.