AnomalyWho made this, and why
There is one person behind this site. It seems fair that you should know who, and what he thinks it is for.
My name is Kyle Clouthier. I live in Petawawa, Ontario, and I write software for a living. I have been interested in the strange and the unexplained since I was a kid, and I never grew out of it. I still watch The Why Files, I have sat through more hours of Ancient Aliens than I would defend in court, and I follow what comes out of Skinwalker Ranch with the same mix of fascination and doubt that most people who watch it will recognise.
It also happened to me
Twice in my life I have woken up unable to move, with something in the room. Both times I was completely awake, or at least everything about it felt more awake than being awake usually does. The first time there was a weight and a presence, the thing people who have had it call the old hag. The second time there was a figure standing there, tall, dark, featureless, with a hat. I did not know at the time that thousands of other people describe a man in a hat. I only knew that I could not move, I could not make a sound, and that I have never in my life been that frightened before or since.
There is a medical name for the state, sleep paralysis, and I am not going to pretend there is not. The body wakes up before the switch that keeps you still during dreaming has flipped back. That explains why I could not move. It is a real explanation and it covers a real part of what happened.
What it does not cover, and what took me years of reading to find out, is why the room is so often occupied by the same visitor. A pressure on the chest and an old woman in Newfoundland. A hat and a long coat in Arizona. The same figure in cultures that had no contact with each other, in centuries that had no way of comparing notes, in languages with their own separate word for it. The paralysis is explained. The consistency of the intruder is not, at least not to the point where anybody can say why the details line up as tightly as they do. That is the part that got hold of me and never let go.
Finding that out took a long time. It was months of following links at two in the morning, through forums and bad websites and archives that were never built to be searched, assembling by hand a picture that any decent instrument would have handed me in a few seconds. I would have given a great deal, in the weeks after the second one, for a page that let me type what happened and told me plainly that it had happened to other people too. Not what it was. Just that I was not the only one. The record now holds 378 accounts of someone being in the room, and 468 of a dark figure. That page exists now. It did not then.
Those two nights are the sharpest of it, but they are not all of it. I have had experiences I would call spiritual, and others I would call supernatural, and I am not going to try to explain any of them to you here. What they left behind was a permanent unwillingness to accept that the unexplained is the same thing as the imagined.
The thing I think about most is everybody who never got as far as I did. Something happens to you that is too frightening or too strange to say out loud. You tell one person, it does not go well, and you stop telling people. The experience does not go anywhere, it just stops being something you are allowed to mention. I would guess that far more of these are sitting quietly in people's heads than have ever been written down anywhere, and every one of those is a piece of evidence that nobody will ever see.
That interest and the work sat apart for a long time. What brought them together was a fairly ordinary observation: this field has enormous archives and almost no instruments. There are hundreds of thousands of accounts sitting in databases that can filter them by date and sort them by state, and not one of those databases can tell you that two strangers on opposite sides of a continent described the same peculiar detail. Everybody is collecting. Nobody is connecting.
The engine came first, and it was for something else
A while back I built a memory system called Neruva. The problem I was trying to solve was a different one: giving software a memory that holds meaning rather than keywords, so it can be asked a question in one form and find what it knows in another, without calling out to a language model every time and without giving a different answer on Tuesday than it gave on Monday. It does that by holding everything it knows in a space of a thousand dimensions and keeping the relationships in a hyperdimensional knowledge graph, which is a mouthful and explained properly on the method page.
It works, and it is useful for what I built it for. But somewhere in the testing it became obvious that the more interesting thing it does is not remembering. It is connecting: taking a very large number of separate, messy, human accounts and finding the threads between them fast enough that you can ask and get an answer while you are still sitting there. That capability was looking for a problem worth having.
This is the problem. 43,684 accounts of things people could not explain, written by people who mostly told nobody else, sitting in archives where nothing was ever going to bring them together.
What I want this to become
The thing I did not have. Somebody has just had the worst night of their life, and instead of months of two in the morning searching they type what happened, in their own words, and find out in seconds whether they are alone. Not a verdict. Not a diagnosis. Just the honest answer to the only question that actually matters that week.
And underneath that, something the field does not have: an instrument rather than a filing cabinet. One that checks the boring explanations first and says so out loud, that can tell twenty five accounts from twenty five witnesses, that shows its reasoning every single time so you can disagree with it, and that treats a case turning out to be a meteor as a good day rather than a bad one. How it works, with the numbers
The rest of the work
Outside this site I build under Clouthier Simulation Labs, which is a sole proprietorship and, in practice, me. The through line in everything there is the same idea you see on this site: a system should be able to show you why it produced an answer, and the answer should come out the same way tomorrow. So alongside Neruva there is a library that makes floating point arithmetic give bit-identical results no matter what order the machine adds things up in, a system for admitting and revoking devices that keeps working when the network is gone, and a tool that turns a claim about a piece of code into a signed certificate anyone can re-run themselves.
There is also FavourBee, which is the odd one out. It is a free mutual-aid network, bilingual, where people ask for and offer help in their own community, and I built and shipped the whole thing myself: the database, the trust graph that stops it being gamed, the interface, the sign-in, the notifications, the French. This site is the same kind of job. I am not a person who writes one layer and hands the rest to somebody else, which is a large part of why a project like this one can exist at all when there is only one of me.
None of that is glamorous and most of it is nothing to do with the unexplained. But it is why this site publishes its misses next to its hits, shows the arithmetic behind every match, and does not ask you to take a connection on faith. That habit came from the day job.
If you are hiring, or you have something hard that needs building, I am open to it. simgen.dev has the detail and LinkedIn has the history, or you can just write to me.
What I am not
I am not a scientist, I am not an investigator, and after everything above I still have no theory about what any of this is. Having been through two of these myself makes me a witness, which is worth exactly as much as anybody else's account here and not a penny more. It does not make me an authority, and I have tried hard not to let it put a thumb on the scale of what the engine finds. When the tool finds something, I publish what it found and how it found it. When it finds nothing, I publish that too. If you know this material better than I do, which many people will, I would rather you told me where I am wrong than nodded along. Tell me.
Kyle Clouthier · Clouthier Simulation Labs, a sole proprietorship registered in Ontario · LinkedIn · the engine is Neruva, also mine