A self-hosted trading desk for Binance. Run strategy bots and grid bots across several markets, backtest with the same rules first, and get daily briefings and trade reviews from an AI partner. Start in paper mode, prove it on testnet, then go live. Nothing changes without your approval.
One WaveTrader instance is one trading desk. The overview shows equity across every running bot, the execution mode, errors, a setup checklist, and a live feed of who did what.

Give each market its own symbol, timeframe, strategy, capital and risk stops. Start them together or one at a time. The engine that places paper fills is the same engine that sends real orders, so nothing changes when you go live.

The grid bot places a ladder of limit orders between two bounds, buying dips and selling the bounce. Before any order goes out, a pre-flight check rejects grids whose spacing would not cover fees.

Backtests run through the same risk logic as the live engine. Stops are tested intrabar against each candle’s high and low, and the circuit breaker halts new entries after a drawdown. A good backtest should then behave the same way live.

The AI partner reads your desk’s actual trades, positions and market regimes. Each morning it writes a briefing, and you can ask it about your own results. It explains what happened and what to watch, and never gives price predictions.

Every fill from every bot goes into an embedded DuckDB warehouse. The analytics view lists each round trip with its P&L, duration, exit reason and the market regime at entry, and can export to CSV.

A nightly analysis compares each strategy’s recent performance with its baseline and with the market regime. When it finds something, it writes a recommendation with the evidence. Nothing changes until someone with the right role approves it.

WaveTrader is multi-user. Owners manage keys and people, traders run bots, and viewers watch. Exchange and AI keys are encrypted at rest, and every meaningful action is written to an audit log.

Live trading needs live keys, the master switch in Settings, and an explicit confirmation. Paper is the default.
Paper simulation and real orders run through the same strategy and risk code. What you tested is what runs.
Stops are tested intrabar and the circuit breaker applies, exactly as the live engine applies them.
The AI explains markets and reviews trades. It never predicts prices, and none of its recommendations take effect without approval.
Positions, fills and grid ladders are saved continuously, and bots resume where they left off after a crash or redeploy.
Self-hosted, with keys encrypted at rest. Your strategies and trade history stay on your own machine.
WaveTrader is a self-hosted algorithmic trading desk for Binance, made by WaveForm Technology Services. It runs strategy bots and grid bots across several markets, backtests them under the same rules first, and adds an AI partner that writes daily briefings and reviews trades.
Binance, in three modes: paper, testnet and live. The engine that places paper fills is the same engine that sends real orders. Live trading needs live keys, the master switch in Settings and an explicit confirmation; paper is the default.
MA crossover (SMA or EMA), RSI mean reversion and MACD, each with its own parameters, plus grid bots with arithmetic or geometric spacing. Every market gets its own stop-loss, trailing stop and drawdown circuit breaker.
No. The AI partner explains markets and reviews trades from your own desk data, and it never gives price predictions. Its optimisation recommendations take effect only after someone with the right role approves them.
Anthropic, Groq, Gemini, OpenRouter, or a local model through Ollama.
On your own server. WaveTrader is self-hosted, exchange keys and other secrets are encrypted at rest, and it ships with Docker, Caddy and systemd configs for a single-server deployment.
We'll walk you through a running desk, from a backtest to a paper bot to the AI briefing, in about fifteen minutes.
A 30‑minute call to understand where you are, where you’re going, and whether we’re the right fit.