We didn't start the-proxies because we love AI. We started it because most of what's being built right now treats AI like a replacement for the person, and we don't think it is.
We're Simone, Frankie and Pancho, with around twenty-five years between us across product, design, construction, business strategy and operations. We were turning specialist knowledge into systems that scale long before “AI” was the word for it. The tools changed. What we believe didn't.
A person with eighteen years of pattern recognition in a domain knows things a generic model never will. The interesting work isn't replacing them. It's giving them a way to be in twenty rooms at once without losing what makes them worth being in the room.
That's what a Helper is. Yours. Not a generic assistant wearing your name. One that works the way you want and still keeps you in the loop. They're called Helpers for a reason.
That's fine for a lot of things. It's a disaster for the work that actually matters. That's the work where the answer depends on who's asking, why, and what they've already tried.
We've seen what happens when a sharp idea gets averaged across too many rounds of feedback. The edges come off, the voice flattens, and what lands is something nobody's really proud of. Generic AI does that same flattening, just faster and at scale. We're not building that.
We build for specificity. We take the things that make a person or a team unrepeatable and turn them into something that scales without flattening what made them worth scaling in the first place.
They're drowning in the work they didn't sign up for. The stuff that grew teeth somewhere along the way and now eats the week:
Most of the people we talk to tell us the same thing. They spend most of their week on the repetitive stuff and wish they had more time for the harder, complex work they're actually best at.
We take it off your plate, in the way you want it done. We focus on the work that's consistent, repeatable, and needs to be accurate every time. That's most of the week for most people, and it's exactly what crowds out the time for the harder calls and the decisions only they can make.
When we're doing it well, the question stops being how do I keep up and starts being what do I want to build next.
If the business model needs people to lose, we're out.
Every one of these uses AI as an amplifier pointed in the wrong direction. The whole reason we exist is that AI multiplies what you point it at, and pointing it at things that quietly hurt people just multiplies the hurt. Some money costs more than it pays.
We've also worked inside enough industries to know how the bad ones rationalise themselves. The conversation always starts with “we're doing it responsibly” and ends with the metrics quietly winning. The simplest way to not end up there is to not start.
We go deep first. The tool, the customer, the constraint, the history of why a problem looks the way it does. Speed comes from understanding the thing, not from skipping it. We co-create with you, and we show up in person when it matters. Most of what's actually useful gets learned in the same room as you.
Of time, of knowledge, of choice. We don't lock people in and we share what we learn, because hoarding it is how you make AI worse for everyone. We want our customers feeling that same kind of freedom too. Free to create, free to grow, free to take the swings they didn't have time for before.
Generic is fine when generic is what's needed. For the rest, we want the version that knows the context, the domain, and the person on the other side of it. We've always been a bit on the edges ourselves. Unorthodox thinkers, the out-of-the-box kind, the people who got called "different" before it was a compliment. It's why we've been good at innovation, and it's why we know capturing the specific way a person or role actually works is where the better outcomes come from.
Our purpose is to help people through this AI transformation in a way that makes AI amplify human knowledge rather than flatten it. That shapes everything else. We work with, and we hire, people trying to make something better. We look for companies with a clear purpose of their own. The won't-do list above is the actual filter, not a marketing line.
AI moves week to week, and most of the people we work with don't have time to chase every shift. We do that work so they don't have to. We test the new models and the patterns nobody's named yet. We bring back what's worth keeping and drop what isn't. Staying on the edge of this is part of the job we signed up for.
There's a difference between output and outcomes, and a lot of consulting confuses the two. Output is the deck, the report, the hours billed. Outcome is whether anything actually changed. We care about the second one. That's why our money is tied to the value we deliver, not the time we spent on it. If the work didn't move the number or ship the thing, we don't think we should be paid for it. We'd rather lose the invoice than win one dishonestly.
This is a forward commitment, not a current scoreboard. As we grow, it will look different in different moments. Pro-bono builds for non-profits whose mission needs the kind of specialist work we do. Free or discounted access to the platform for groups who can't pay but should have it. Straight money to causes that back people who need a leg up.
We'll publish what we did each year, starting from year one. The point is to have the receipts, not to use them as marketing.
This commitment is inspired by broader calls within the AI industry for companies to direct a share of AI-driven gains back to society. OpenAI's Industrial Policy for the Intelligence Age proposes a Public Wealth Fund — seeded in part by AI companies themselves — to give every citizen a stake in AI-driven economic growth. Their CEO has separately suggested a 2.5% annual equity contribution from companies above a certain valuation. We think waiting for governments to mandate it misses the point. We'd rather start now — well, soon actually. We need to get a few legal ducks in a row first, but the intent is locked in.
We don't see AI as the villain or the saviour. It's a tool, the most powerful one we've had, and the people who get to shape what it does over the next decade have an obligation to be honest about what it actually is.
We're trying to be some of those people.
Simone, Frankie & Pancho