TL;DR
Get self-care and calm essentials delivered free — and shop member deals
- Fast, free delivery on millions of items
- Access to Prime Big Deal Days deals on October 6–7
- Prime Video, Amazon Music and more included
Thorsten Meyer AI has announced Forezai · TradingAgents, a research framework that uses multiple AI agents to simulate a trading desk. The project is framed as experimental software for structured market analysis, not as financial advice or a trading recommendation.
Thorsten Meyer AI has announced Forezai · TradingAgents, a research framework that models a trading firm as a group of specialized AI agents, including analysts, opposing bull and bear researchers, a trader and a risk manager with veto power.
The confirmed release is part of the Forezai Markets family and follows Polybot, described in the source material as a single AI forecaster. TradingAgents shifts the design from one model producing one market view to a structured process in which different agents gather signals, argue competing cases and pass any proposed action through risk review.
According to Thorsten Meyer AI, the framework is available at forezai.com/tradingagents.html. The project is described as local-first and provider-agnostic, meaning roles can be run on owned compute and, in principle, assigned to different swappable models.
The source material repeatedly states that TradingAgents is not financial advice, not a recommendation to trade or invest, and not a claim of profitability. It says automated trading carries a substantial risk of loss, including total loss of capital, and that market and trading-software access may be regulated or restricted depending on jurisdiction.
TradingAgents — a firm made of agents
A single model is an overconfidence machine. So this isn’t one AI — it’s a whole desk: analysts, a bull and a bear who argue, a trader, and a risk manager who can say no.
Not financial, investment, legal or tax advice; not a recommendation or solicitation to trade, invest or use any software. Forezai · TradingAgents is an experimental research framework, provided “as is” without warranty of accuracy or profitability. Trading and automated trading carry a substantial risk of loss including total loss of capital; past or backtested performance does not indicate future results. Market and trading-software access is regulated or restricted in some jurisdictions — you are solely responsible for compliance with applicable law. Consult a licensed professional before any financial decision. Produced with AI assistance under human editorial oversight; independent commentary, the author’s own views. Product and company names are trademarks of their respective owners; mention does not imply endorsement.
Agent Debate Meets Market Risk
The release matters because it applies a wider trend in AI system design to financial research: using multiple agents with assigned roles instead of relying on one model’s single answer. In the TradingAgents design, a bullish case and a bearish case are both developed before a trader agent proposes an action.
The risk manager role is central to the claim being made by the project. The source material says the risk function can vet, size or veto a proposed decision, with a conservative default that may result in no trade. That makes the framework less a claim about beating markets and more a test of whether structured disagreement can reduce overconfident model output.
For readers following AI tools in finance, the project is relevant because it highlights the gap between market analysis software and deployable trading systems. The announcement presents TradingAgents as a research framework whose reasoning can be recorded, not as a finished investment product.
As an affiliate, we earn on qualifying purchases.
Polybot Now Has A Desk
The announcement is the Day 14 entry in Thorsten Meyer AI’s 19-day Built in Public series. The source material says the prior Markets entry focused on Polybot, a single AI forecaster comparing one estimate with one market price.
TradingAgents is positioned as the second half of that Markets layer: Polybot represents the lone forecaster, while TradingAgents represents a simulated firm. The broader operator portfolio described in the source material includes 18 products across content, decision, platform, markets, defense and diagnostic categories.
The design also extends an idea referenced by the author as a council approach: multiple agents are used to surface competing interpretations before a decision is made. In this case, that structure is aimed at financial research, where confident but wrong conclusions can have direct monetary consequences.
As an affiliate, we earn on qualifying purchases.
No Track Record Disclosed
The source material does not provide live trading results, audited performance data, independent testing or details on how the framework performs across market regimes. It also does not specify which models should run each role, what data providers are required or how users should evaluate output quality.
It is also unclear how much work is needed to adapt the framework for regulated environments. The announcement warns that market access and trading-software use can be restricted in some jurisdictions, but it does not provide legal guidance for any specific country or broker setup.
financial market simulation software
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Testing Will Define Use
The next step for readers and developers is inspection rather than deployment: studying the agent workflow and testing the framework in non-production settings. Any use connected to real money would require separate legal, technical and financial review by qualified professionals.
The Built in Public series is also scheduled to continue beyond Day 14, with TradingAgents completing the Markets family alongside Polybot, according to the source material.
As an affiliate, we earn on qualifying purchases.
Key Questions
What is Forezai · TradingAgents?
It is a research framework that simulates a trading desk using multiple AI agents, including analysts, bull and bear researchers, a trader and a risk manager.
Is TradingAgents financial advice?
No. The source material states that it is not financial advice, not a recommendation to trade or invest, and not a guarantee of accuracy or profit.
What license does the project use?
Forezai · TradingAgents is developed privately and is not publicly available.
How is it different from a single AI forecaster?
The framework assigns different roles to different agents so that market views can be challenged before any proposed action reaches risk review.
Can it be used for automated trading?
The source material warns that automated trading carries substantial risk, including total loss of capital, and that market access may be regulated or restricted. Anyone considering financial use should consult qualified professionals.
Source: Thorsten Meyer AI
Fall Picks
fall essentials
As an affiliate, we earn on qualifying purchases.
