<div class="row mt-5">
        <div class="col-md-4 mb-4">
            <div class="p-4 h-100" style="background: var(--cyber-bg-card); border: 1px solid var(--cyber-border); border-radius: 12px;">
                <h3 class="h4 mb-3" style="color: var(--cyber-accent);">1. Execution Infrastructure</h3>
                <p>Our algorithms run on a low-latency, containerized execution stack designed for 24/7 reliability.</p>
                <ul class="list-unstyled">
                    <li class="mb-2">✅ <strong>BaFin-Compliant Architecture.</strong> Engineered by alumni of the German banking sector.</li>
                    <li class="mb-2">✅ <strong>Kubernetes-Orchestrated.</strong> Isolated containers with dedicated resource limits.</li>
                    <li class="mb-2">✅ <strong>Geographic Redundancy.</strong> Distributed nodes across high-availability zones.</li>
                </ul>
            </div>
        </div>
        <div class="col-md-4 mb-4">
            <div class="p-4 h-100" style="background: var(--cyber-bg-card); border: 1px solid var(--cyber-border); border-radius: 12px;">
                <h3 class="h4 mb-3" style="color: var(--cyber-accent);">2. Multi-Layered Circuit Breakers</h3>
                <p>We don't just set stop-losses; we manage systemic risk across the entire book.</p>
                <ul class="list-unstyled">
                    <li class="mb-2">✅ <strong>Hard Stop-Losses.</strong> Deterministic, hard exit prices defined at entry.</li>
                    <li class="mb-2">✅ <strong>Drawdown Kill-Switches.</strong> Execution pauses if historical drawdown thresholds are breached.</li>
                    <li class="mb-2">✅ <strong>Flash-Crash Detection.</strong> Real-time volatility monitoring to protect against fat-tail events.</li>
                </ul>
            </div>
        </div>
        <div class="col-md-4 mb-4">
            <div class="p-4 h-100" style="background: var(--cyber-bg-card); border: 1px solid var(--cyber-border); border-radius: 12px;">
                <h3 class="h4 mb-3" style="color: var(--cyber-accent);">3. Quantitative Validation</h3>
                <p>Our strategies are stress-tested against historical anomalies and randomized paths.</p>
                <ul class="list-unstyled">
                    <li class="mb-2">✅ <strong>Walk-Forward Optimization.</strong> Validation on unseen data to prevent over-fitting.</li>
                    <li class="mb-2">✅ <strong>Black Swan Stress Tests.</strong> Simulations of 2008, 2020, and crypto liquidity events.</li>
                    <li class="mb-2">✅ <strong>Monte Carlo Simulations.</strong> Testing robustness against randomized trade sequences.</li>
                </ul>
            </div>
        </div>
    </div>

    <div class="row mt-5 pt-4">
        <div class="col-lg-10">
            <h2 class="mb-4">The "Black Box" Problem</h2>
            <p>We believe in <strong>Explainable AI</strong>. Every signal used by our strategies is based on measurable, economic intuition — whether it's mean reversion in FX or sentiment-drift in equities. We provide full documentation for every model, allowing institutional allocators to understand the <em>why</em> behind the PnL.</p>
            <div class="skin-box mt-4">
                <span class="skin-icon">🛡️</span>
                <span class="skin-text"><strong>Robustness mandate:</strong> We optimize for <strong>Sortino</strong> and <strong>Calmar</strong> ratios, prioritizing capital preservation over raw equity growth.</span>
            </div>
        </div>
    </div>
</div>