A Transformer that reads the regime
Tensor’s custom architecture conditions forecasts on whether the market is trending up, trending down or ranging. The same price pattern can mean different things in different market states.
Trading Evolved
A new standard in
deep-learning directional trading.
The market is non-stationary and chaotic. Adapt every decision based on a dynamic system.
A premium strategy series developed by Tensor, with regime-aware intelligence from entry to exit.
01 / The philosophy
Markets are non-stationary and chaotic. Archimedes is designed to adapt its decisions as the relationships between price, liquidity and volatility change.
Archimedes pairs a proprietary, regime-aware time-series Transformer with an adaptive, unsupervised exit engine. It reads trend, range and volatility before deciding how to enter, size, hold and exit a position.
Each preview strategy is a portfolio of complementary long and short sub-strategies, spanning trend, mean-reversion and volatility regimes, with upgraded shorting capability relative to earlier engines such as Hackworth.
02 / The engine
From reading the market to managing the last fill, Archimedes brings prediction, portfolio construction and execution into one adaptive system.
Tensor’s custom architecture conditions forecasts on whether the market is trending up, trending down or ranging. The same price pattern can mean different things in different market states.
A proprietary, lightweight unsupervised model continuously re-evaluates open positions. Across the preview strategies, 72–83% of exits were model decisions rather than stop-loss triggers.
Curated inputs span price action, market microstructure, volatility and cross-asset context, giving the model a broad view of the state it is trading.
Each strategy combines directional, mean-reversion and volatility sleeves, with balanced long/short exposure and improved shorting capability relative to earlier engines such as Hackworth.
A deep reinforcement learning execution layer uses dynamic TWAP to slice orders, accelerating when liquidity is deep and backing off when it is thin.
Every sub-strategy is evaluated independently through standalone Sharpe, parameter perturbations, trade bootstraps, matched-control exit benchmarks and year-by-year stability checks.
03 / The preview strategies
Four strategies form the core series — BNB, ETH, SOL and DOGE. Two belong to Archimedes Pi, behind the insignia: XMR and ZEC. All six were evaluated under an identical, deliberately demanding protocol.
| Strategy | Series | Cumulative return | Sharpe ratio | Max drawdown | Win rate | Profit factor | Trades |
|---|---|---|---|---|---|---|---|
| BNB | Archimedes | +1,509% | 2.89 | −21.8% | 59.5% | 1.50 | 2,184 |
| ETH | Archimedes | +1,536% | 3.54 | −16.3% | 49.6% | 1.80 | 1,287 |
| SOL | Archimedes | +1,525% | 3.31 | −15.4% | 55.9% | 1.43 | 2,297 |
| DOGE | Archimedes | +4,763% | 3.27 | −26.5% | 48.9% | 1.40 | 1,998 |
| XMR | Archimedes Pi | +5,513% | 3.87 | −21.8% | 65.4% | 1.66 | 3,692 |
| ZEC | Archimedes Pi | +28,295% | 4.17 | −22.3% | 45.6% | 1.91 | 2,601 |
Frozen sub-strategy portfolios, January 2024 – September 2026 · $1.0M initial research equity · 300% target gross exposure · baseline scenario. Figures cover the full window: the research phase runs through July 2026, with data from August 2026 onward marking the live test run. Backtest results; not a guarantee of future performance.

Sharpe ratioReturn earned per unit of risk. Above 1 is good, above 2 is excellent; the series spans 2.89–4.17.
Maximum drawdownThe worst peak-to-trough dip the account ever experienced. Smaller is calmer.
Win rateThe share of trades closed profitably.
Profit factorDollars won per dollar lost: 1.50 = $1.50 earned per $1.00 given back.

Frozen portfolio backtest, Jan 2024 – Sep 2026 · $1.0M initial equity · 300% target gross exposure · baseline conditions. Exit benchmark: 1-hour horizon, 1,743 matched control events. Historical results do not guarantee future performance.

Frozen portfolio backtest, Jan 2024 – Sep 2026 · $1.0M initial equity · 300% target gross exposure · baseline conditions. Exit benchmark: 1,065 matched control events. Historical results do not guarantee future performance.

Frozen portfolio backtest, Jan 2024 – Sep 2026 · $1.0M initial equity · 300% target gross exposure · baseline conditions. Exit benchmark: 1,645 matched control events. Historical results do not guarantee future performance.

Frozen portfolio backtest, Jan 2024 – Sep 2026 · $1.0M initial equity · 300% target gross exposure · baseline conditions. Historical results do not guarantee future performance.
04 / Archimedes Pi
A strictly limited sub-series
The pinnacle of the Archimedes series — state-of-the-art deep-learning strategies taken to their furthest extreme, metered in Points and strictly limited.
XMR · ZEC · ACCESS STRICTLY LIMITED

Frozen portfolio backtest, Jan 2024 – Sep 2026 · $1.0M initial equity · 300% target gross exposure · baseline conditions. Exit benchmark: 2,870 matched control events. Historical results do not guarantee future performance.

Frozen portfolio backtest, Jan 2024 – Sep 2026 · $1.0M initial equity · 300% target gross exposure · baseline conditions. Historical results do not guarantee future performance.
05 / Access
Premium alpha is scarce, and putting it to work takes research, compute and trading capacity. Points connect the value you create on AlphaNet with the strategies you choose to run.
Caps apply per wallet to each individual strategy, not to the combined allocation across a tier. Strategy availability, capacity and account risk limits still apply.
Every tier earns 1 Point per dollar deployed per day, including leveraged deployment. Daily earning follows the lowest deployed amount in the previous seven days, so a brief one- or two-day increase does not lift it.
Archimedes uses 2 Points per $1 per day against 1 earned, a net cost of 1. Archimedes Pi uses 4, a net cost of 3.
To open or add to a Tier 2 or Tier 3 position, your Points balance must be sufficient to last at least seven days afterwards.
The app warns at seven, three and one days remaining, followed by a seven-day grace period. If the shortfall continues, Tier 3 stops first, then Tier 2. Tier 1 is never stopped by Points, and your balance never goes below zero.
There is no minimum deployment requirement for Tier 2 or Tier 3 today. Allocations remain subject to strategy capacity and account risk controls, and a full strategy queues its waitlist by Points balance.
Request Tier 2 and Tier 3 access through the AlphaNet team. Points levels, allocation limits and referral rates are reviewed quarterly.
Every figure in this brochure is backed by a complete benchmark dossier — standalone signal Sharpe ratios, parameter sweeps, trade bootstraps and matched exit benchmarks — available on request.
Read the Points policy ↗06 / The proving ground
A strong portfolio result is only the beginning. Each selected sub-strategy signal was evaluated independently.
Each of the 39 sub-strategy signals must clear Sharpe 1.0 on its own.
Every model exit is compared with matched “what if we held” controls, measuring the value of closing the position.
141–158 neighboring parameter settings per strategy; 90–99% remained reliable.
2,000 re-draws of each signal’s trade history; 93–100% probability of a positive outcome.
Every sub-strategy re-scored year by year, covering 2024, 2025 and 2026 to date.