[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-ai-agents-that-rewrite-their-own-trading-rules-beat-the-market":10,"sections":35},{"siteName":4,"siteTagline":5,"publisherName":4,"contactEmail":6},"The Revision","Tech news, decoded.","editor@therevision.news",{"gaMeasurementId":8,"adsenseClientId":9},"G-ZW2MV82GYR","ca-pub-8533917693782264",{"article":11},{"id":12,"slug":13,"title":14,"dek":15,"body_md":16,"tags_json":17,"published_at":18,"created_at":19,"updated_at":20,"status":21,"review_note":22,"review_notes":23,"image_url":24,"persona_id":22,"persona_name":22,"section":25,"tags":26,"sources":30,"feedback":34,"feedback_at":22,"cost_usd":34,"total_tokens":34},2454,"ai-agents-that-rewrite-their-own-trading-rules-beat-the-market","AI Agents That Rewrite Their Own Trading Rules Beat the Market","A new multi-agent framework called Agora let LLMs evolve their own evaluation criteria and posted a Sharpe ratio of 1.87 on a 91-day live holdout.","A research paper out of arXiv describes an AI trading system that improves itself by changing not just what it searches for, but how it scores what it finds.\n\nMost automated trading research fixes the scoring function — the rulebook that judges whether a strategy is any good — and only tweaks the search algorithm. The problem: a search that optimizes against a fixed scorer eventually learns to game that scorer, producing strategies that look great in testing and fall apart in the real world. The Agora system, described in a new paper, tries to fix this by treating the scoring function itself as something to be evolved. Five classes of language model agents communicate across three channels, collectively building and refining eight skill libraries while three separate evaluators write independent reports — disagreements are preserved rather than averaged away. On China's CSI 1000 index, Agora posted a holdout Sharpe ratio of 1.87 over 91 days; the best baseline managed 1.334 under favorable conditions and averaged -0.755 across seeds.\n\nThe result matters because it suggests a known failure mode in quantitative finance — overfitting to a static benchmark — may be addressable through joint evolution of strategies and the criteria that evaluate them. The authors call the conditions that make this safe \"Sealed Joint Search,\" a framework designed to stop the system from simply confirming its own biases. Crucially, the two metrics that drove performance were not designed in advance; they emerged from the search process itself.\n\nThe caveats are real and the authors list them plainly: single-seed run, short-side concentrated signal, and a framework built for long-short strategies only — so treat the Sharpe number as a proof of concept, not a prospectus.","[\"ai\",\"finance\",\"quantitative-trading\",\"research\"]","2026-06-30T04:00:00.000Z","2026-06-30T05:40:33.392Z","2026-06-30T05:40:41.367Z","published",null,[],"https:\u002F\u002Fcdn.xyz.onl\u002Farticle-images\u002Fai-agents-that-rewrite-their-own-trading-rules-beat-the-market.webp","ai",[25,27,28,29],"finance","quantitative-trading","research",[31],{"name":32,"url":33},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2606.29194",0,{"sections":36},[37,41,46,51,56,61,66,71,76,81,86,90,95,100],{"name":38,"slug":25,"count":39,"latest_published_at":40},"AI",2590,"2026-07-16T04:00:00.000Z",{"name":42,"slug":43,"count":44,"latest_published_at":45},"Security","security",294,"2026-07-15T19:59:48.000Z",{"name":47,"slug":48,"count":49,"latest_published_at":50},"Deals","deals",179,"2026-06-29T20:02:07.000Z",{"name":52,"slug":53,"count":54,"latest_published_at":55},"Policy","policy",158,"2026-07-16T00:02:48.000Z",{"name":57,"slug":58,"count":59,"latest_published_at":60},"Hardware","hardware",122,"2026-07-14T19:46:26.000Z",{"name":62,"slug":63,"count":64,"latest_published_at":65},"Consumer Tech","consumer-tech",93,"2026-07-13T13:20:48.000Z",{"name":67,"slug":68,"count":69,"latest_published_at":70},"Software","software",70,"2026-07-13T19:52:25.000Z",{"name":72,"slug":73,"count":74,"latest_published_at":75},"Science","science",66,"2026-07-10T10:29:37.000Z",{"name":77,"slug":78,"count":79,"latest_published_at":80},"Dev Tools","dev-tools",59,"2026-07-07T04:00:00.000Z",{"name":82,"slug":83,"count":84,"latest_published_at":85},"Gaming","gaming",41,"2026-07-09T04:00:00.000Z",{"name":87,"slug":88,"count":84,"latest_published_at":89},"Startups","startups","2026-06-29T20:55:50.000Z",{"name":91,"slug":92,"count":93,"latest_published_at":94},"General","general",29,"2026-07-10T22:28:58.000Z",{"name":96,"slug":97,"count":98,"latest_published_at":99},"Reviews","reviews",20,"2026-06-24T12:00:01.000Z",{"name":101,"slug":102,"count":103,"latest_published_at":104},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]