Live equity scoring

An AI‑native
hedge fund built on systematic models.

Our system continuously scores 1,000+ equitiesuniverseAbout 1,000 globally listed equities, monitored continuously. Well beyond the 30-50 names a traditional analyst can cover. on fundamentals, behavior, and regime shifts. Medium-term alpha, not intraday noise. +151.8% since inception vs. +41.3% VT Index. Paper trading results; past performance does not guarantee future results.

Apply to Allocate View demo See the track record
scoring / live
online
Track Record

Exceptional alpha. Verified through paper trading. Ready for capital.

Live since June 2024. Twenty-four months of systematic trading, fully logged and auditable.

Paper trading disclosure: All performance figures on this page reflect simulated paper-trading results from June 2024 through June 2026. They are not live fund returns. Past performance does not guarantee future results.

Cumulative return vs. global benchmarks

Quant AI VT Index VTI EFA
Jun 2024 to Jun 2026
Quant AI cumulative return vs benchmarks Jun 2024 to Jun 2026 Quant AI paper trading cumulative return plus151.8 percent compared with VT Index plus41.3 percent, VTI plus42.0 percent, and EFA, from June 2024 through June 2026. +150% +100% +50% 0% -25% Jun'24 Jan'25 Jul'25 Jan'26 Jun'26 +151.8%
Sources: Quant AI internal paper-trading ledger (Jun 2024–Jun 2026); VT and VTI cumulative returns from Vanguard ETF data. Benchmarks shown for comparison only.
Cumulative returns Jun 2024 to Jun 2026
SeriesCumulative return
Quant AI (paper)+151.8%
VT Index+41.3%
VTI+42.0%
Cumulative Return
0.0%
Jun 2024 to Jun 2026
vs. VT Index +41.3%
2025 Net Return
0%
vs. VTI +15.91% (2025)
≈ 54% annualized alpha
Sharpe Ratio
0.00
Sortino 3.78
Max drawdown 13.4%
Information Ratio
>1.9
Tracking error vs. VT Index 7.1%
Excess return vs. benchmark

* Paper trading results, June 2024 through June 2026. Benchmark: VT (Vanguard Total World Stock ETF), tracking the FTSE Global All Cap Index. Past performance does not guarantee future results.

The System

A coordinated model stack, running continuously across the full universe.

Inference costs have fallen sharply. Multi-model scoring across 1,000+ names is now practical at fund scale.

01

Full universe coverage

Agents covering fundamentals, sentiment, regime, and flow score about 1,000 global equities in real time. A single analyst typically covers 30-50 names.

02

Medium-term horizon

Buy and sell signals for 6-18 month holds. We focus on fundamental dislocations and behavioral mispricings, not microstructure noise dominated by HFT.

03

Continuous monitoring

Systematic scoring across the full universe, 24/7. Every decision logged, every position auditable.

Platform

Quant AI runs on a proprietary Large Financial Model. The first use case is in production.

The model ranks public equities for alpha, including long and short candidates across 1,000+ names with continuous, auditable scores. Built for capital markets from the ground up.

Why Now

Incumbents move slowly. We already ship.

What changed

100× cheaper inference
Model costs per token have fallen by two orders of magnitude in 24 months. Running scores across 1,000+ securities is now economical.
Stronger models
Current models can weigh competitive dynamics, fundamentals, and regime shifts, not just price patterns.
Real-time fundamentals
The system updates scores as fundamental news arrives, across the full coverage universe.

What stops incumbents

Legacy architecture
Bloomberg and FactSet were built for data delivery, not model-driven scoring. Retrofitting takes years.
Innovator's dilemma
$24K Terminal seats fund a $10B model. Why cannibalize the data licensing flywheel?
Org inertia
Banks still run COBOL from 1959. Commonwealth Bank: $750M and five years just to modernize core.

Why us

18-24 month head start
In production today, with extensive backtesting and live paper trading complete.
RL infrastructure
Twelve months or more to build properly. It requires live trading data and compounds once deployed.
Workflow depth
Customer feedback improves the models. Funds that embed the stack face real switching costs.

If algorithms beat discretion for thirty years,
why are we still analyzing stocks with spreadsheets?

Systematic, model-driven equity selection at institutional scale.

Get in touch