How ATLAS was born

ATLAS wasn't born in a boardroom. It was born in my apartment in Madrid, with an Excel spreadsheet open and the accumulated frustration of four hours of work to analyze a single stock.

This is the story of how we got to the version you're using today. Not the pitch-deck version — the real one.

The first 40%

I started investing when I was eighteen. Two hundred, three hundred dollars. What I had. The experience was, literally, like a casino. I won, lost, won again. Without understanding what was happening.

I stopped for a while. I came back during my time at IE University studying economics in Madrid — but this time with a process.

The process was simple and rigorous: for any stock I was considering, I'd build three things. An intrinsic value model to understand what the business was worth. A read on the macro environment. And a deep dive on the company itself: management, financials, news, what analysts were saying.

Three pillars. Each one documented in Excel. Each one took three or four hours per stock.

I did this work for months. I generated a 40% return in one year — an exceptional result by any benchmark. But something was bothering me.

The problem I couldn't let go

Not everyone could do what I was doing. Not because they lacked intelligence — because they lacked time. Because they lacked the obsessive interest that made a four-hour stock analysis feel enjoyable rather than grueling. Because they lacked the technical knowledge to build the models.

The information existed. The math worked. But the process was a barrier — high enough that most people simply went without it and made their investment decisions based on whatever they had last read on the internet.

That bothered me. It bothered me enough that I couldn't let it go.

The conversations that lit the fuse

I started talking to friends about how they invested. The answers disturbed me. One invested based on what he had seen in the news. Another described a vague sense of which sectors felt good right now. A third admitted he mostly just bought whatever had been going up lately.

None of them had a process. None of them had a framework. They had information — noise, mostly — and they were converting it into decisions by way of instinct, anxiety, and peer pressure.

And they were losing money.

I thought: there has to be a platform that fixes this. Something that takes all the relevant inputs and produces not just more information, but an output — a number — that tells you what your actual odds are.

I looked for it. I couldn't find it. Every tool in the market gave you data. Charts, news feeds, analyst ratings. Data everywhere, in every format, from every angle. What none of them gave you was what I had always built myself in Excel: a probability. A single honest answer to the question every investor is actually asking, which is not "what do analysts think about this stock?" but "if I buy this today, what are the odds I make money?"

August 2025

In August 2025 I started building the tool. The first version was called SAR — my initials. It was, honestly, bad. It sent long reports that were hard to read and hard to act on. It was technically functional. Practically, useless.

I killed it.

I rebuilt it from scratch. This time the architecture was different: not a reporting tool but a simulation engine. The core mechanic was Monte Carlo — the same probabilistic modeling technique used by quantitative hedge funds, risk management departments at major banks, and engineering teams planning missions to Mars.

I had not studied this formally. I learned it by doing. Iterating, testing against historical data, refining the inputs, verifying the outputs.

Zero lines of code

There's one important detail that deserves particular attention: when I started, I had no programming background. None. I couldn't read code. I certainly couldn't write it.

Today, ATLAS has more than 100,000 lines of code.

The distance between those two points — from zero to one hundred thousand — was covered in approximately eight months, using artificial intelligence as both teacher and collaborator. Building, testing, breaking, and rebuilding. Seven days a week. In a Madrid apartment. Alone.

What I take from the road

Before ATLAS there were other things. Four or five projects. MyGym, a portable home gym concept I launched at eighteen during the pandemic. XALTE, a gourmet salt brand in Spain. Each one taught me something different about building, about market timing, about when to keep pushing and when to let go.

Not all of them survived. Most didn't. And that's fine.

I entered 2025 carrying all of that learning. Six years of trying things, losing things, learning things. And an idea I couldn't let go of.

That's what most founder stories don't tell you: the project that works is almost always the one that came after the ones that didn't. ATLAS isn't my first attempt. It's my sixth.

What I believe

What I believe, at the core, is that ordinary people are being robbed. Not maliciously, necessarily — but structurally. The tools for making genuinely informed investment decisions have always existed. The mathematical frameworks, the simulation techniques, the probability models — they're not new. Institutional investors have been using them for decades.

None of that reaches the retail investor. None of it reaches the person with two thousand dollars in a brokerage account who's trying to decide whether to buy Apple or hold Tesla. They get news feeds and price charts and star ratings. They get information without interpretation. They get everything except the one thing they actually need, which is an honest probability.

My initial instinct — and I'll tell you this without ceremony — was to keep the model private. I had something that worked. I could have used it exclusively for myself, made money quietly, and told no one.

I decided not to.

I decided to make the full platform available from day one, with no feature gating. And to offer a seven-day free trial with no credit card required.

There's something genuine in the egalitarian impulse behind that — a belief that financial intelligence should not be a luxury good, and that the democratization of these tools matters. I want people to build their financial lives better. I want people to stop losing money on vibes and start making decisions with actual math behind them.

ATLAS is that. It's the result of six years of wanting something like it to exist, eight months building it alone, and the conviction that the answer to "what are my actual odds?" should be available to anyone who wants to know.

Welcome.

Santiago Avalos · SAR