Introduction

ATLAS is a quantitative investment platform for people who don't want to guess. This documentation explains how the engine works, how to use every feature, and how to build your own integrations.

The core of ATLAS is a Monte Carlo simulation engine that runs against every stock on the NYSE, NASDAQ, and AMEX. When you ask "what's the chance NVDA hits +5% in 21 days?", the engine runs 10,000 independent simulated paths for NVDA's price over the next 21 days — each calibrated to NVDA's own historical behavior, current volatility, and any relevant macro or news signals — and counts how many of the 10,000 paths ended at or above +5%. That fraction is your probability.

Everything else — the portfolio builder, the morning brief, recovery odds, replacement suggestions — is built on top of that engine.

What this documentation covers

Getting started introduces core concepts and walks through your first simulation. Read this if you're new.

The engine is the technical heart: methodology, calibration, what signals feed the model, and how accuracy is measured. Read this if you want to trust the numbers.

Simulations / Portfolios / Tracking cover the three main product areas in depth.

API documents the public endpoints for integrating ATLAS into your own tools.

Reference includes the full track record, an FAQ, and the product changelog.

Not financial advice

ATLAS is a calculation and analysis tool. It does not give financial advice, does not execute trades, and does not guarantee any outcome. Markets can move in ways no model captures. Use ATLAS probabilities as one input to your decisions, not the whole decision.