On-Chain Quant Trading — Backtest, Optimize and Deploy Strategies
whale.ag is the on-chain quant platform for perpetual traders. Author strategies in a declarative DSL, walk-forward optimize on real Hyperliquid fills, and deploy with one click. Built for solo quants, prop desks and crypto-native funds.
Why whale.ag
- Declarative DSL: write strategies the way you describe them — no Python boilerplate, no event loops.
- Backtest on real on-chain fills: realistic slippage, exchange fees and funding rates baked in.
- Walk-forward optimization: automatically re-fit parameters on rolling windows; protect against curve fitting.
- One-click deploy: same DSL runs in backtest and in production, on Hyperliquid (and HIP-3) standard perps.
- Programmatic risk: per-strategy stop-loss, position caps, daily volume caps and equity guardrails.
- Compare strategies side by side; promote the winner via feature flag.
From idea to live strategy in 5 minutes
1. Author strategy in DSL
Open /quant/dsl and write your strategy declaratively. Conditions, sizing, exits, risk caps.
2. Backtest
Hit Run. The engine replays your DSL over real on-chain fills with realistic slippage, fees and funding.
3. Optimize
Walk-forward optimize parameters on rolling windows; review out-of-sample performance, Sharpe and max drawdown.
4. Deploy
Promote the winning configuration. whale.ag spins up an agent wallet and runs the strategy on live Hyperliquid markets.
5. Monitor
Track equity curve and drift in real time. Pause, adjust or kill the strategy at any moment.
Frequently asked questions
Do I need to know Python to use whale.ag quant?
No. whale.ag uses a declarative DSL designed for traders, not engineers. You describe entry/exit conditions, sizing rules and risk caps — the engine takes care of execution.
What data does the backtester use?
Real on-chain fills from Hyperliquid (and other supported DEXs), with realistic slippage, exchange fees and funding rate accrual baked in. No synthetic data.
Can I deploy a strategy to Hyperliquid HIP-3 markets?
Yes. The same DSL targets standard perps and HIP-3 builder-deployed markets (stocks, commodities, indices).
How does whale.ag prevent over-fitting?
Walk-forward optimization with rolling in-sample/out-of-sample windows; out-of-sample Sharpe is what gets reported. You can also penalize parameter complexity automatically.
What does it cost to deploy a quant strategy?
Current rates are published at docs.whale.ag. There is no monthly subscription and no extra charge for using the quant module.
Get started
whale.ag is the on-chain quant platform for perpetual traders. Author strategies in a declarative DSL, walk-forward optimize on real Hyperliquid fills, and deploy with one click. Built for solo quants, prop desks and crypto-native funds.
Open the strategy editor