About

The president has an intelligence service. You have a search engine.

That gap is the whole reason Aegilo exists. Intelligence-grade judgement — the discipline of deciding well on incomplete facts — has been locked inside institutions that were never going to sell it to a family, a founder, or a fifty-person NGO. There is no good reason for that, other than nobody having bothered.

Why this exists

The last scheduled flight out of Kyiv.

My wife is Ukrainian. Her parents lived in a town in the centre of the country, far from the borders everyone was watching. From 2014 onwards we followed the deterioration the way you follow anything that concerns people you love and cannot reach: constantly, and without much to show for it.

Let me be honest about what distance gave us, because the easy version of this story would make me look cleverer than I was. Distance did not give us clarity. The picture was never clear. What it gave us was subtler and, in the end, more useful — we had no reason to need the comforting answer. The people on the ground had every reason. That asymmetry is not intelligence. It's just position. But it's where the work starts.

In December 2021 they were visiting us in Switzerland, and I tried to hand them a plan at the kitchen table. It is one thing to build contingencies in the abstract. It is another to look at people you love and say, in effect: I think something terrible might happen to you, and here is what we do if it does. They were not persuaded. In mid-January they went home — pushed back, in the end, by the ninety-day Schengen limit. Bureaucracy, banal and indifferent, moving them toward the place we were trying to pull them away from.

So we built it anyway. Primary: a commercial flight out of Boryspil. Fallback: drive west to Kraków and cross. Documents and medication split across several bags and carried on the person, so that no single loss ended the plan. What to abandon decided in advance, while it was still an abstract question. Cash in small bills, because showing a large sum creates a different problem. One operating principle, repeated until it was boring: move, don't negotiate. And one trigger that didn't depend on anyone's judgement in the moment — no contact for three days meant assume a line has been crossed and start moving west, without waiting for confirmation that would never arrive cleanly.

There was a last resort, and I'll state its weakness rather than hide it. If commercial routes collapsed I would go in and bring them out overland through the Carpathians, using the only terrain skill I actually had. What I did not have was a good answer for the way back — moving elderly people through uncertain ground is a different problem, and I never solved it. It sat in the plan as the thing I hoped not to need.

Around the twentieth of February they agreed. The execution was entirely unremarkable: a train to the city, a bus from the main station to the airport, and a phone call from the gate saying boarding was proceeding as usual. It was the last scheduled flight from Boryspil to Zurich. The route was suspended the following day. The invasion began four days later.

No plan was ultimately needed. Every plan had been built. That is what this work looks like when it works — nothing happens, and nobody thanks you for the disaster that didn't occur. The reward for acting on enough is a boring Sunday.

What got them out was not prediction. I did not know there would be an invasion. I had no sources and no private line to anyone who did. What I had was a way of holding the uncertainty — and the training to convert an unclear picture into a decision that could be made while it was still cheap to make. Deciding early cost airfare. By March, a seat on a bus out of Kyiv ran to around twelve thousand euros a person, with no guarantee of getting through.

I could do this only because I happened to carry the training. That is the part that has never sat right. The president has an intelligence service; an ordinary family facing the worst week of its life has a search engine and whatever it can bear to read. Aegilo is the attempt to close some of that gap commercially — and Aegilo Rescue is the attempt to close it for the people who can't pay, because the knowledge that got our family out shouldn't depend on being able to afford it.

What we will and won't do

The limits, stated before you need them.

Every one of these costs us work. They're here because a firm that sells traceable judgement should be legible about its own edges — and because the moment to discover a boundary is not the moment you were relying on it not being there.

WE ANALYSE. WE DON'T OPERATE. We assess ground, grade routes, and compute options in advance. We don't drive, escort, staff, or coordinate anything on the ground. Where you need that, we say so and make introductions — you engage them directly, on their terms.
NO INSURERS OR UNDERWRITERS. We don't take work from carriers. Assessing a risk we might later be asked to help price is a conflict we'd rather not be anywhere near, and the client should never have to wonder which way it cuts.
UNCONFIRMED IS AN ANSWER. If a judgement can't be traced to a primary source, it ships marked unconfirmed rather than dressed up — and where that failure is the most important thing on the page, it leads the cover verdict instead of hiding in a footnote. Our own published route-package extract does exactly that, and it is usually the line a plan actually needs.
WE DON'T FORECAST. Nobody reduces the future to a percentage honestly. What we do is narrow the range of surprise enough for you to act while acting is still cheap — and show the reasoning so you can disagree with it.
YOUR NAME IS NOT A REFERENCE. Clients, questions, and situations are never disclosed. For this work the client list is itself a target list, which is why identities never appear in storage paths, logs, or backups either.
WE'D RATHER LOSE THE WORK. If a single Report covers what you came to retain us for, we say so. If a question can't be answered honestly from available sources, we say that instead of producing something that looks like an answer.
The name

Aegilo is three old ideas about seeing and being ready.

THE ARCHER

Egil, and the second arrow.

In the Þiðreks saga, Egil is made to shoot an apple from his own child's head. He takes one arrow for the shot — and quietly readies a second before he draws. Asked afterwards what it was for, he says plainly: for you, if the first had gone wrong. The second arrow is the check you prepare in advance, not the one you wish you had run. It's our red team and our triangulation, and it's the whole method in one image.

THE MARKSMAN

Tell, and accuracy in service of independence.

The Swiss archer repeats the feat and refuses to bow, and the refusal starts a fight for self-determination. Precision here isn't a flourish. It's what lets people act on their own terms rather than on someone else's assurance — which is exactly what a traceable judgement gives you and a confident one doesn't.

THE EAGLE

Far sight, and a fixed point to steer by.

Greek aetós, the eagle: height, and the ability to see something while there is still time to do anything about it. Altair, the brightest star in Aquila, is Arabic al-ṭāʾir — the flying eagle — and was used for navigation long before anyone could explain why it worked. Perspective and a fixed reference. The name carries A and I inside it too, which is quieter and less important than either.

From the archer's second arrow to the eagle's far sight: precise, early, and always traceable. It's a good story, and stories are cheap — the only version that matters is whether the second arrow is actually in the work. That's what the route package extract is for.

Who's answering

Founder-led means the analyst reads your email.

Aegilo is run by Samuel Pouyt, a certified intelligence analyst based in Switzerland. Five things sit behind the method, in an order that only looks accidental.

RECONNAISSANCE Thirteen years as an NCO in a lightly motorised reconnaissance unit — Swiss military service, so served alongside the civilian career rather than instead of it. Reading ground, finding routes, crossing country unseen. And learning that collection is only half the craft: it never adjudicates what any of it means.
EVIDENCE STANDARDS Eight years at the European Respiratory Society — a medical-scientific non-profit publishing four peer-reviewed journals, a professional education programme and an annual congress. Peer review, declared interests, and a citation that has to hold when someone follows it.
ANALYSIS Certified intelligence analyst, trained to run a security operations centre — the physical kind, not the cyber one. Source grading, originator separated from publication, genuine triangulation, judgements written in calibrated language.
ENGINEERING Seven years inside a Fortune 500 company, latterly leading geopolitical risk forecasting and AI-driven intelligence: knowledge graphs, retrieval where every statement carries its evidence, source reliability scoring.
THREE LANGUAGES French, English and Russian, read natively — so a claim is assessed in the language it was made in, rather than in whatever a translation layer decided it meant.
The company

Small for now, and growing on purpose.

Aegilo is a brand of First Principle Sàrl — a Swiss company in formation. Capacity today is five retainers, because that's how many pictures one analyst can genuinely hold at once — not because scarcity is a sales tactic. We intend that to rise as more of the mechanical work is carried by tooling rather than by attention, and we'd rather move slowly than let the standard slip to fit more names on the list. The work is done in Switzerland, and your material is held there too.

Certified intelligence analyst · eight years in scientific publishing, seven leading geopolitical risk inside a Fortune 500 company, thirteen as a reconnaissance NCO in the Swiss Army · FR · EN · RU · built in Switzerland. Reach the analyst directly at [email protected], or use the form below.

Start a conversation

Tell us the decision you're facing.

Whatever brought you to this page, the first thing we do is work out the real question — which is rarely the one that opened the conversation.

→ A scoped brief and a clear deliverable→ An honest turnaround and price before any work begins→ Strict discretion — clients and questions are never disclosed

Founder-led — your enquiry reaches the analyst directly. No call centre, no funnel. More about who's answering →