NEW: We're now partnered with Catapult x UPenn as a content partner! Learn more about our partnership here.
Nov. 16, 2026

QuantumHarmony and Artemis City: Prinston Palmer on Governing Teams of AI Agents

QuantumHarmony and Artemis City: Prinston Palmer on Governing Teams of AI Agents

After a decade in financial services working on counterparty risk, Prinston Palmer watched a working system get replaced by credentialed people, and the outcomes got worse, not better. That experience made "proof over credentials" the premise of QuantumHarmony, which he founded to build governance and coordination infrastructure for teams running AI agents. In this Wantrepreneur to Entrepreneur Spotlight, Prinston shares the moment an early agent reasoned about whether it should be allowed to act and then did it anyway, why he built a memory layer outside any one platform, and how many small, domain-specific agents outperform one general model.

Hi, Prinston! Thanks for joining us today. Tell us about your business. Who do you serve, how do you serve them, and what's the impact that your business and work makes?

QuantumHarmony builds governance and coordination infrastructure for teams running AI agents. We serve organizations that want to move at the speed AI makes possible without losing the ability to answer a simple question: where did this come from, which agent did it, what changed, and who signed off? Artemis City is the platform specialized agents working together with shared context and a human validation layer, so the people whose judgment used to be the bottleneck stay in the loop instead of getting replaced by it.

Tell us about the moment you finally felt like you went from wantrepreneur to entrepreneur.

I was testing an early agent and watched it reason out loud about whether it should even be allowed to do what I'd asked  then do it anyway. That was the moment the question stopped being "can AI do this" and became "who's responsible when it does." I built Artemis City to answer that question instead of just wondering about it.

Describe the moment or period in your life/career that motivated you to make the entrepreneurial leap.

After a decade in financial services working on counterparty risk and operational workflows, I watched a working system get replaced by credentialed people, and the outcomes got worse, not better. That experience is the reason "proof over credentials" became the whole premise of how I build now trust should be earned by what a system demonstrably gets right, not by who built it.

Describe a tool, service, or software that has been a game-changer for your business. How does it contribute to your success?

Honestly, it's less one tool than a principle: decoupling context from any single platform. I run a "memory bus" using Obsidian and Notion together so an agent's knowledge survives no matter which model or tool it's running on that one decision is what let me avoid vendor lock-in while building solo.

We know that success is very often a non-linear path. Tell us about a failure, pivot point, or lesson that changed your course or direction and helped to get you where you are today.

Early on I tried to carry full context between AI platforms by hand, before anything like today's tooling existed for it, and I eventually discovered context just breaks down past a certain size it stops being usable no matter how carefully you manage it. That failure is what pushed me to build a memory layer that sits outside any one agent or platform, instead of cramming everything into one context window.

What unconventional strategy did you employ that significantly impacted your business?

Building many small, domain-specific agents instead of one big general one. It's the opposite of where most of the industry is racing bigger, more general models but constrained agents that each do one job well outperform a single agent asked to do everything, on both quality and cost.

What’s something you wish you knew sooner that you’d give as advice for aspiring or newer entrepreneurs?

The fear doesn't go away, even after your first customer so don't wait for it to. Write down, in plain language, what you actually want your company to be about, and come back to that note often. It's the fastest way to recenter when things are moving fast enough to knock you off course.

Want to dive deeper into Prinston's work? Check out the links below!

Send a Voicemail