How I Learned to Run Operations With AI as My Partner

Running operations across an entire company means living at the intersection of every function. In any given day I move between product strategy, business development, legal and compliance, finance, and the dozens of small operational fires that never make it onto an org chart. The job was never to be the deepest expert in each of those areas, it’s to keep them connected, moving, and accountable.

For years that meant a state of near-permanent context-switching: holding too much in my head, jumping between vocabularies and stakeholders, and quietly accepting that in the churn, some details would slip.

That changed when I stopped treating AI as a novelty and started treating it as infrastructure.

I want to be clear about what that does and doesn’t mean. I haven’t handed my judgment to a machine, and I’m not interested in operations-by-autopilot — especially not in a regulated, high-stakes industry where a dropped detail has real consequences. What I’ve built instead is a working partnership: AI handles the synthesis, the first drafts, and the connective tissue, and I stay where I add the most value, making decisions, setting direction, and owning outcomes. Here’s what that looks like in practice.

From meeting chaos to a living source of truth

The single biggest unlock has been turning the constant stream of meetings into something structured and accountable. Across a busy month, I sit in conversations spanning every department, each one generating decisions, dependencies, and follow-ups that historically lived in scattered notes and good intentions.

Now I let AI do what it’s genuinely good at: ingesting a mountain of unstructured notes and pulling out the signal. I can take weeks of discussions and, in minutes, produce a clean, owner-by-owner action tracker — who committed to what, by when, and what’s still open. What used to be an afternoon of reconstructing “wait, what did we actually decide?” is now a few focused minutes of review. The value isn’t just the time saved. It’s that nothing falls through the cracks, and the whole team operates from one shared, current picture instead of everyone’s slightly different memory of the same meeting.

A drafting partner that keeps pace with the work

Operations runs on documents. Strategy decks, one-pagers, internal memos, process guides, executive briefings, the deliverables are endless, and the blank page is the slowest part of producing any of them.

I almost never start from zero anymore. I bring the raw thinking, the strategy, the framing, the points that matter, and AI produces a structured first draft I can react to. Reacting to something is so much faster than conjuring it. A presentation that once took the better part of a day now takes an hour or two, and the time I reclaim goes back into the substance: sharpening the argument, pressure-testing the logic, making sure the thing actually says what I need it to say. The AI gets me to a credible draft; I make it right.

This matters most when I’m translating across audiences. The same underlying initiative has to be explained one way to engineers, another to a commercial buyer, another to a compliance reviewer, and another to the executive team. Being able to quickly reshape a single set of ideas for very different readers, without rebuilding from scratch each time, has quietly become one of my most-used capabilities.

Research on demand

A surprising amount of operations leadership is really applied research. What are competitors doing? How do peers structure a given function? What does a particular market, program, or regulatory pathway actually require? These questions used to mean either an open-ended afternoon of digging or a note to circle back later and “later” doesn’t always come.

AI has compressed that cycle dramatically. I can frame a question precisely, get a structured first pass on the landscape, and then spend my energy on the part only I can do: judging what’s credible, what’s relevant to us specifically, and what it means for our next move. It hasn’t replaced rigor or verification, I still check what matters before I act on it. but it’s removed the friction that used to keep good questions from getting asked at all.

A thinking partner, not an answer machine

The use case I value most is the hardest to put on a slide. Some of my best operational decisions have come out of essentially talking a problem through, laying out a messy situation, asking for the counterargument, stress-testing a plan before I bring it to the team.

Used this way, AI is less a tool than a tireless thought partner that never gets impatient with a half-formed idea. It will tell me what I might be missing, surface the considerations I’d glossed over, and play devil’s advocate on demand. I don’t outsource the decision, I sharpen it. The judgment, the accountability, and the final call stay firmly mine. But I’m reaching those calls faster and with more of the angles considered.

Guardrails are the whole point

None of this works without discipline, and in an industry as regulated as mine, the discipline is non-negotiable. I’m deliberate about what information goes where, I keep sensitive material out of contexts it doesn’t belong in, and I treat AI output as a strong first draft to be verified, never as a final authority to be trusted blindly. The technology is a force multiplier for good operational judgment; it is not a substitute for it. Operators who forget that distinction don’t move faster. They just make their mistakes faster.

That’s the mindset I’d encourage anyone in a similar seat to adopt: the goal isn’t to do less thinking. It’s to spend your thinking on the things that actually require you.

What it actually changed

The honest bottom line is capacity. I’m effectively running the workload of a larger team without one, covering more ground, across more functions, with fewer details lost in the gaps. The low-value work that used to consume my days, reconstructing notes, formatting documents, chasing down what was decided has largely receded, and what’s left is the work I was actually hired to do: strategy, decisions, and keeping a complex organization aligned and in motion.

We talk a lot about AI replacing roles. My experience has been the opposite. It’s made me a more effective version of the operator I already was, one with more reach, more leverage, and more room to lead. In a function defined by wearing every hat at once, that’s not a small thing. It’s the difference between surviving the firehose and getting ahead of it.

If you run operations and you’re still treating AI as a curiosity, my advice is simple: stop. Treat it as part of how you work. Build the guardrails first, keep your judgment in the driver’s seat, and then let it carry the weight it’s built to carry. The job is hard enough. There’s no prize for doing the parts a machine could do for you.