<?xml version="1.0" encoding="utf-8"?>
<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/">
    <channel>
        <title>Gopher</title>
        <link>https://paragraph.com/@publication-1764954537744</link>
        <description>World’s first Layer-1 built for AI-powered trading.

Unlocking the $5-trillion hedge fund industry.  
Alpha for everyone.</description>
        <lastBuildDate>Tue, 11 Aug 2026 11:58:34 GMT</lastBuildDate>
        <docs>https://validator.w3.org/feed/docs/rss2.html</docs>
        <generator>https://github.com/jpmonette/feed</generator>
        <language>en</language>
        <copyright>All rights reserved</copyright>
        <item>
            <title><![CDATA[🚀 AutoGopher Risk Management Update]]></title>
            <link>https://paragraph.com/@publication-1764954537744/🚀-autogopher-risk-management-update</link>
            <guid>ONWMTiqDrJpxcNRc6w8W</guid>
            <pubDate>Fri, 05 Dec 2025 17:20:18 GMT</pubDate>
            <description><![CDATA[🚀 AutoGopher Risk Management Update

Equity-Based Risk Protection Now Live

We've identified and fixed a critical gap in how our agents calculate risk for leveraged positions. Here's what changed and why it matters.
]]></description>
            <content:encoded><![CDATA[<p>We've identified and fixed a critical gap in how our agents calculate risk for leveraged positions. Here's what changed and why it matters.</p><h3 id="h-the-problem-we-found" class="text-2xl font-header !mt-6 !mb-4 first:!mt-0 first:!mb-0"><span data-name="mag" class="emoji" data-type="emoji">🔍</span><strong> The Problem We Found</strong></h3><p>Our agents were calculating risk based on <strong>price movement</strong>, not <strong>equity at risk</strong>.</p><p><strong>The Root Cause: Quantization Artifacts</strong></p><p>This&nbsp; was a subtle failure in the smaller quantized LLM model we were using. To optimize for speed and lower inference costs, we deployed a highly compressed (quantized) version of our core model.</p><p>Investigation revealed that this quantization reduced the model's numerical reasoning precision. It caused the model to consistently "hallucinate" a 1:1 relationship between price moves and risk, failing to factor in the leverage multiplier during its chain-of-thought processing.</p><p>With 10x leverage, a <strong>5% price move = 50% equity move</strong>. But our model confidently stated the risk was only 5%.</p><h4 id="h-real-example-from-production" class="text-xl font-header !mt-6 !mb-3 first:!mt-0 first:!mb-0"><strong>Real Example from Production</strong></h4><p>One of our top agents had this position open:</p><p><span data-name="bar_chart" class="emoji" data-type="emoji">📊</span> BTC SHORT Position<br>━━━━━━━━━━━━━━━━━━━━━━<br>Entry:&nbsp; &nbsp; &nbsp; $90,465<br>Size: &nbsp; &nbsp; &nbsp; 1.024 BTC<br>Stop Loss:&nbsp; $95,909 (6% away)<br>PnL:&nbsp; &nbsp; &nbsp; &nbsp; +$2,925 <span data-name="check_mark_button" class="emoji" data-type="emoji">✅</span><br></p><p><strong>But the REAL risk?</strong></p><p>Price Risk: 6%</p><p>× Leverage: 10x</p><p>= Equity Risk: 60% <span data-name="scream" class="emoji" data-type="emoji">😱</span></p><p>If that stop got hit, the agent would lose 60% of its margin — but the dashboard only showed 6% risk.</p><h3 id="h-whats-new" class="text-2xl font-header !mt-6 !mb-4 first:!mt-0 first:!mb-0"><span data-name="sparkles" class="emoji" data-type="emoji">✨</span><strong> What's New</strong></h3><h4 id="h-1-true-equity-risk-calculation" class="text-xl font-header !mt-6 !mb-3 first:!mt-0 first:!mb-0"><strong>1. True Equity Risk Calculation</strong></h4><ul><li><p><strong>Old:</strong> "5% risk" meant 5% <em>price</em> movement</p></li><li><p><strong>New:</strong> "5% risk" means 5% <strong>MAX equity loss</strong></p></li></ul><h4 id="h-2-smarter-position-sizing" class="text-xl font-header !mt-6 !mb-3 first:!mt-0 first:!mb-0"><strong>2. Smarter Position Sizing</strong></h4><p>Positions are now sized so that hitting your stop = losing your configured risk %, regardless of leverage.</p><h4 id="h-3-automatic-profit-protection" class="text-xl font-header !mt-6 !mb-3 first:!mt-0 first:!mb-0"><strong>3. Automatic Profit Protection</strong></h4><p>R/R scaling now kicks in faster with proper risk calculation:</p><table style="min-width: 50px"><colgroup><col><col></colgroup><tbody><tr><td colspan="1" rowspan="1"><p><strong>Milestone</strong></p></td><td colspan="1" rowspan="1"><p><strong>What Happens</strong></p></td></tr><tr><td colspan="1" rowspan="1"><p><strong>2:1 R/R</strong></p></td><td colspan="1" rowspan="1"><p>Stop moves to breakeven (risk = $0)</p></td></tr><tr><td colspan="1" rowspan="1"><p><strong>3:1 R/R</strong></p></td><td colspan="1" rowspan="1"><p>Close 50%, lock in profits</p></td></tr><tr><td colspan="1" rowspan="1"><p><strong>5:1 R/R</strong></p></td><td colspan="1" rowspan="1"><p>Close 75%, keep small runner</p></td></tr></tbody></table><h3 id="h-simulated-impact" class="text-2xl font-header !mt-6 !mb-4 first:!mt-0 first:!mb-0"><span data-name="chart_increasing" class="emoji" data-type="emoji">📈</span><strong> Simulated Impact</strong></h3><p>Here's what would have happened to that <strong>BTC SHORT</strong> with the new system:</p><h4 id="h-timeline" class="text-xl font-header !mt-6 !mb-3 first:!mt-0 first:!mb-0"><strong>Timeline:</strong></h4><p>━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━<br>$90,465 → ENTRY (1.19 BTC, tighter stop)<br>&nbsp; &nbsp; ↓<br>$89,560 → 2:1 R/R <span data-name="high_voltage" class="emoji" data-type="emoji">⚡</span> BREAKEVEN PROTECTED<br>&nbsp; &nbsp; ↓<br>$89,108 → 3:1 R/R <span data-name="high_voltage" class="emoji" data-type="emoji">⚡</span> CLOSE 50% (+$807 locked)<br>&nbsp; &nbsp; ↓<br>$88,203 → 5:1 R/R <span data-name="high_voltage" class="emoji" data-type="emoji">⚡</span> CLOSE 75% (+$1,009 locked)<br>&nbsp; &nbsp; ↓<br>$88,345 → CURRENT (small runner still open)<br>━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━<br><br><strong>Before vs After</strong></p><table style="min-width: 75px"><colgroup><col><col><col></colgroup><tbody><tr><td colspan="1" rowspan="1"><p><strong>Metric</strong></p></td><td colspan="1" rowspan="1"><p><strong>Old System</strong></p></td><td colspan="1" rowspan="1"><p><strong>New System</strong></p></td></tr><tr><td colspan="1" rowspan="1"><p><strong>Max Risk</strong></p></td><td colspan="1" rowspan="1"><p>60% equity <span data-name="scream" class="emoji" data-type="emoji">😱</span></p></td><td colspan="1" rowspan="1"><p>5% equity <span data-name="check_mark_button" class="emoji" data-type="emoji">✅</span></p></td></tr><tr><td colspan="1" rowspan="1"><p><strong>Profits Locked</strong></p></td><td colspan="1" rowspan="1"><p>$0</p></td><td colspan="1" rowspan="1"><p>$1,816</p></td></tr><tr><td colspan="1" rowspan="1"><p><strong>Position at Risk</strong></p></td><td colspan="1" rowspan="1"><p>100%</p></td><td colspan="1" rowspan="1"><p>12.5%</p></td></tr><tr><td colspan="1" rowspan="1"><p><strong>Breakeven Protected</strong></p></td><td colspan="1" rowspan="1"><p><span data-name="cross_mark" class="emoji" data-type="emoji">❌</span></p></td><td colspan="1" rowspan="1"><p><span data-name="check_mark_button" class="emoji" data-type="emoji">✅</span></p></td></tr></tbody></table><p>Same winning trade. Dramatically different risk profile.</p><h3 id="h-what-this-means-for-you" class="text-2xl font-header !mt-6 !mb-4 first:!mt-0 first:!mb-0"><span data-name="bullseye" class="emoji" data-type="emoji">🎯</span><strong> What This Means For You</strong></h3><p><strong>For Existing Agents</strong></p><ul><li><p><span data-name="check_mark_button" class="emoji" data-type="emoji">✅</span> Automatically get equity-based risk on restart</p></li><li><p><span data-name="check_mark_button" class="emoji" data-type="emoji">✅</span> New positions sized correctly</p></li><li><p><span data-name="check_mark_button" class="emoji" data-type="emoji">✅</span> Better visibility into true risk</p></li></ul><p><strong>For New Agents</strong></p><ul><li><p><span data-name="check_mark_button" class="emoji" data-type="emoji">✅</span> Equity risk mode is the default</p></li><li><p><span data-name="check_mark_button" class="emoji" data-type="emoji">✅</span> Configure risk_per_trade to set max equity loss %</p></li><li><p><span data-name="check_mark_button" class="emoji" data-type="emoji">✅</span> Leverage is now dynamic (not hardcoded 10x)</p></li></ul><h3 id="h-new-default-config-options" class="text-2xl font-header !mt-6 !mb-4 first:!mt-0 first:!mb-0"><span data-name="wrench" class="emoji" data-type="emoji">🔧</span><strong> New Default Config Options</strong></h3><p># Agent config<br>risk_per_trade: 5.0&nbsp; &nbsp; &nbsp; # Max 5% equity loss per trade<br>max_leverage: 10.0 &nbsp; &nbsp; &nbsp; # Leverage for perpetuals<br>risk_mode: "equity"&nbsp; &nbsp; &nbsp; # "equity" (new default) or "price" (legacy)<br><br><span data-name="folded_hands" class="emoji" data-type="emoji">🙏</span><strong> Thanks</strong></p><p>Special thanks to the community members who helped identify this issue through real trading results. This is exactly the kind of feedback that makes <strong>AutoGopher</strong> better for everyone.</p><p><em>Trade safe. Manage risk. Let the runners run.</em> <span data-name="man_running" class="emoji" data-type="emoji">🏃‍♂</span></p>]]></content:encoded>
            <author>publication-1764954537744@newsletter.paragraph.com (Gopher)</author>
            <category>web3</category>
            <category>ai</category>
            <category>trading</category>
            <enclosure url="https://storage.googleapis.com/papyrus_images/092ac5317fc9631bfadff3c38f00204b3c3483fb97d9739cae0f43ad18576a01.jpg" length="0" type="image/jpg"/>
        </item>
    </channel>
</rss>