{
  "id": 13823865,
  "title": "Why I stopped trading and started tendering with AI",
  "url": "https://urgent.news/2026/10/11/why-i-stopped-trading-and-started-tendering-with-ai",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-11T23:33:00.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/sam_hiotis_117598dbfa3ac2/why-i-stopped-trading-and-started-tendering-with-ai-2o8i"
  },
  "original_language": "en",
  "account": "For years, the author dedicated themselves to algorithmic trading, investing significant time and resources in building systems that would outperform the market. However, the constant struggle against changing market conditions and the threat of unforeseen events proved to be exhausting. Feeling like Sisyphus, pushing an unending boulder uphill, the author eventually discovered an alternative approach: tendering with AI.\n\nRather than attempting to predict market movements, the author shifted their focus towards responding strategically to market changes. This article shares the author's journey from a frustrated trader to an AI-powered tendering architect, detailing technical challenges faced and explaining why this approach is more sustainable and potentially profitable in the long run.\n\nTraditionally, tendering referred to a bidding process, but in this context, it involves identifying market inefficiencies and strategically deploying capital to exploit them. The author emphasizes that tendering is not about beating the market but about being a sophisticated participant, leveraging AI to assess risk and opportunity, and making calculated bids for short-term advantages.\n\nThe author adapted their existing trading infrastructure to transition from traditional trading algorithms to tendering agents. They simplified a moving average crossover trading algorithm and created a tendering logic function that assigns a score based on the moving average crossover and volatility. Higher volatility signals a greater potential opportunity and risk, increasing the tender score for buy signals. A risk_tolerance parameter determines when trades (tenders) are executed.\n\nTo further optimize the system, the author incorporated Reinforcement Learning using a Proximal Policy Optimization (PPO) agent. This agent, trained on historical market data, dynamically adjusts the risk_tolerance and weighting of the tender score components. The RL agent's interaction with the tendering logic is demonstrated through pseudocode, which starts with retrieving market state features, such as price, volume, and volatility, then predicts an adjusted risk_tolerance value. With this new risk_tolerance, the tender function is executed, returning a trade signal and tender score.",
  "summary": "Why I Stopped Trading and Started Tendering with AI For years, I chased the dragon of algorithmic trading. Backtesting strategies, optimizing parameters, chasing slippage… the whole nine yards. I poured countless hours (and frankly, too much money) into building systems that promised to consistently outperform the market. And for a while, some did. But the constant arms race, the changing market…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}