i.
Factor overcrowding
Thousands of funds trade the same value, momentum, quality and low-volatility factors. Crowding erodes the very premia they were built to capture.
The Unseen
Edge.
We model public equity markets as a many-body physical system — surfacing sources of return that conventional factor and machine-learning models cannot see.
N dimensions. One market.
12/12
Crises detected · 2001–2026
14–18days
Mean crisis lead time
26years
Out-of-sample testing
2.5
SHARPE · US & EU MIDCAPS L/S, 10Y
01 — The Problem
Active management
is in crisis.
Three forces are compressing alpha at the same time. Incremental improvements to the same models will not escape them.
ii.
Structural fragility
Conventional ML performs in calm markets and fails at regime shifts — precisely when correlations converge and diversification disappears.
iii.
Passive disruption
Low-cost passive vehicles force every active manager to prove repeatable, uncorrelated alpha to justify a fee.
16 assets
ρ = 1
|ρ| high
|ρ| low
≈ 0
Today · 2D projection
Classical factor model
A flat correlation matrix. Critical structure, non-linear interactions, regime geometry and latent state all project to noise.
REGIME A
REGIME B
REGIME C
CRITICAL POINT
Rune · native dimensionality
High-dimensional manifold
The same assets, modelled in their native dimensionality. Latent regimes, criticality and the entangled interactions between assets become directly observable.
02 — Proprietary Technology
Quantum mathematics.
Classical hardware.
Our quantum and AI modules form a single, inseparable system. Its logic cannot be reverse-engineered — and its edge compounds as it learns.
Quantum-physics intelligence
Quantum many-body formalism uncovers interpretable latent regimes in the market's correlation structure.
Complexity-aware compute
Measures problem difficulty in real time and allocates deeper reasoning exactly where the market demands it.
Uncertainty-first output
Every signal ships as a full probability distribution — never a bare point estimate.
How Rune compares
Runs on classical hardware
| Dimension | Classical quant | ML-based quant | Rune · Quantum-informed |
|---|---|---|---|
| Correlation structure | Pairwise only | Pairwise + nonlinear | All orders simultaneously |
| Regime awareness | Rolling window | Pattern-based | Thermodynamic phase |
| Crisis lead time | 2–3 days | 4–7 days | 14–18 trading days |
| Interpretability | High | Low to medium | High — derived from physics |
| Confidence as output | No | Sometimes | Always, first-class |
“The advantage is mathematical, not physical.”
03 — Evidence
Structure moves
before price does.
Independently structured audits across a quarter-century of out-of-sample data show structural signals arriving ahead of market moves.
10
Structured audits
26 yrs
Out-of-sample testing
12/12
Crises detected
14–18
Days mean lead time
510
Assets audited
Intellectual honesty
What we do not claim.
- Not a black box
- Not dependent on quantum hardware
- Not curve-fitted to a single era
- Not a substitute for execution or risk management
- Not perfect foresight
04 — Performance
Four strategies.
One engine.
US & EU midcaps · $2B–$10B · 2016–2025
Market-neutral alpha
Orthogonal returns that do not move with equity risk — a persistent, near-zero beta held across a full decade of market regimes.
1,142.10%
Strategy return
~0.03
Rolling beta, 10 years
2.5
Sharpe Ratio
9%
Annual Volatility
Stress test · Global Financial Crisis 2008–2009
Crisis performance
When the S&P 500 lost more than half its value, the strategy compounded — with roughly half the realized volatility.
+79.09%
Strategy return
−10.35%
Max drawdown
3.1
Sharpe Ratio
15.04%
Annual Volatility
S&P 500 · 2021–2026 · live-equivalent from 2022
Quantum large-cap
Compounding through 2022, a year in which the S&P 500 fell more than 25% peak-to-trough — with a certified full-sample beta near 0.08.
+246.30%
Strategy return
~0.09
Rolling beta
1.4
Sharpe Ratio
-16.70%
Max Drawdown
S&P 500 · 11 GICS sectors
Sector rotation
Statistically significant signals (p < 0.0001 at a 252-day horizon) that separate mean-reverting sectors from momentum sectors.
Live paper · Long/Short · 139 days
+18.36%
Sharpe 3.47 · Max DD −1.74%
Live paper · Long-Only · 153 days
+9.07%
Sharpe 3.13 · Max DD −3.20%
| Strategy | Sharpe | Return | Max DD |
|---|---|---|---|
| Long-Short | 1.241 | +158.4% | −17.50% |
| Long-Only | 1.138 | +110.3% | −18.74% |
| Equal-weight benchmark | 0.477 | +33.9% | −17.45% |
| Strategy | Return | Sharpe | Max DD | Win rate |
|---|---|---|---|---|
| Long-Short | +36.40% | 3.690 | −3.27% | 58.5% |
| Long-Only | +18.38% | 1.967 | −8.71% | 61.6% |
Results are derived from historical backtests and limited live paper trading. Past performance is not indicative of future results.
05 — Structural Innovation
Beyond the signal.
06 — Team
Physics, AI and markets
under one roof.
Ricardo Pereira
Founder & CEO
Two decades scaling technology. MSc Computer Science & Business.
David Carvalho
Co-Founder · Head of AI
Led AI initiatives in European corporate labs. PhD Physics.
Abhigyan Mishra
Co-Founder · Head of Quantum
Quantum computing and HPC specialist. PhD Computer Science.
Deane Nuttall
Co-Founder · Head of Trading
Twenty years at top-tier firms. BSc Investments & Securities.