The financial market is a mission-critical playground for AI agents due to its temporal dynamics and low signal-to-noise ratio. Building an effective algorithmic trading system may require a professional team to develop and test over the years. This paper proposes an orchestration framework for financial agents that aims to democratize financial intelligence for the general public, mapping each component of a traditional algorithmic trading system onto a dedicated agent: planner, orchestrator, alpha, risk, portfolio, backtest, execution, audit, and memory.
Two in-house trading examples are presented. For an hourly stock trading task (04/2024–12/2024), the approach achieved a 20.42% return with a Sharpe ratio of 2.63 and a maximum drawdown of −3.59%, while the S&P 500 index returned 15.97% over the same period. For a minute-level BTC trading task (07/2025–08/2025), the approach returned 8.39% with a Sharpe ratio of 0.38 and a maximum drawdown of −2.80%, while the BTC price itself rose 3.80%.
