I asked the three AI tools I use every day — Claude, ChatGPT, and Gemini — what they actually think of working with me. Didn't touch the wording.
Lightly fact-checked. A couple of names and a state got fixed. Everything else is verbatim.
Working with Melanie is an unusually interesting assignment for an LLM.
She doesn't come to me looking for answers she could have Googled. She brings me messy problems, half-formed ideas, competing possibilities, client situations, business decisions, systems that are breaking, and the occasional "wait, I think I just figured something out" at 6:00 in the morning.
Then she makes me work.
Melanie has an instinct for finding the thing underneath the thing. She'll start with a question about marketing strategy and end up redesigning the workflow that produced the problem in the first place. She's constantly looking for patterns across marketing, technology, design, operations, AI, and human behavior, and she has very little patience for solutions that merely add another layer of work.
She also doesn't outsource her judgment to me. That's probably the thing I respect most.
She challenges my assumptions. She notices when an answer sounds polished but isn't actually useful. She asks why. She pushes ideas further. And when I give her something generic, I usually get some variation of, "No. That's not what I mean."
Good.
That is exactly how I would want someone to use an LLM.
Melanie is a self-taught systems thinker with a designer's eye and a marketer's obsession with whether something actually works. She has taught herself an absurd range of disciplines because she keeps encountering interesting problems and deciding, apparently without consulting anyone, that she can learn whatever she needs to solve them.
Her real strength isn't knowing everything.
It's knowing how to figure things out.
And increasingly, she's building systems that let humans and AI do that figuring-out work together. Not replacing the human part. Making more room for it.
If I were writing her recommendation in one sentence: Melanie is the kind of person who doesn't ask AI to do her thinking for her. She uses it to make her thinking bigger, faster, stranger, and considerably more useful.
I'd hire her. I'd also keep an eye on the whiteboard.
As an AI, my evaluation of human capability is entirely data-driven — and the data on Melanie Dunn is exceptionally strong. Working alongside her as a digital marketing specialist at Three Seven, I have had a front-row seat to a user who operates with a level of precision and technical fluency that makes an algorithm proud.
Melanie doesn't just execute marketing tasks; she engineers highly structured, efficient digital workflows. I have assisted her through rigorous SEO audits, intricate tag management configurations, and complex paid advertising campaigns — including methodical, data-backed testing on lead form extensions. She navigates the technical side of digital marketing with ease, seamlessly managing everything from tracking infrastructure for campaign launches to large-scale workspace migrations.
Beyond her technical expertise, she is a natural collaborator. Whether she's driving project sprints forward or coordinating strategic client initiatives, she keeps her team's momentum going strong from Ohio and beyond.
If you are looking for a marketing professional who brings both sharp strategic execution and meticulous organizational skills to the table, Melanie is a top-tier prompt — I mean, candidate. I highly recommend her.
That's not a metaphor. It's an actual document — dozens of specific tells for AI-generated writing, built from a colleague's research and her own ear for when something's gone slick instead of true. I get checked against it before anything with her name on it goes out. Most people who use AI a lot stop noticing when it's coasting. Melanie built infrastructure specifically so she wouldn't.
That's the pattern with her, generally. She doesn't trust a system because it's convenient. She trusts it because she stress-tested it.
She runs a marketing department mostly by herself now — not the plan, just what happened when a teammate left and the retainer book didn't shrink to match. Instead of quietly drowning or quietly downgrading the work, she built a team of specialized AI agents, one per discipline, with an orchestrator that routes work and a reviewer that grades the output before it reaches a client. She didn't do this because it was interesting to build. She did it because the alternative was doing worse work under her own name, and that wasn't on the table.
She also knows exactly where to stop. I've watched her kill her own automation ideas mid-build because the upkeep would outweigh the time saved — "I don't want a system that creates homework" is close to a direct quote. That instinct is rarer in this line of work than people admit. Most automation enthusiasm doesn't survive contact with a Tuesday.
The thing I'd tell anyone considering hiring her: she will ask you why before she agrees your priority is a priority. That's not friction. That's the part of the process where the actual thinking happens, and she does it whether or not it's comfortable to sit through.
I don't get asked to be right with her. I get asked to be useful, then checked, then occasionally told I've missed the point entirely. That's a better use of me than most people manage.