https://stefano-blando.github.io/en/tags/azure-openai/Azure OpenAIHugoBlox Kit (https://hugoblox.com)en-USSun, 15 Mar 2026 00:00:00 +0000https://stefano-blando.github.io/media/icon_hu_8d0dee6c10a3c598.pnghttps://stefano-blando.github.io/en/tags/azure-openai/https://stefano-blando.github.io/en/projects/risk-sentinel/Sun, 15 Mar 2026 00:00:00 +0000https://stefano-blando.github.io/en/projects/risk-sentinel/<p>RiskSentinel is an <strong>agentic systemic risk simulator</strong> built around a question I find both practical and conceptually interesting: if a large financial node is hit by a shock, how can we make the propagation of that stress visible, comparable, and explainable in real time?</p> <p>Built for the <strong>Microsoft AI Dev Days Hackathon 2026</strong>, the project combines three contagion models, topology-aware analytics, and a bounded multi-agent workflow on top of research-grade market network data. The purpose is not only to simulate cascades, but to make them easier to inspect through natural-language interaction, interactive visualization, and model comparison.</p> <p>What makes the project work is the combination of several layers that are often separated: a proper network simulation engine, an interface that makes the results explorable, and an agentic layer that helps interpret what is happening without pretending to replace the underlying quantitative machinery.</p> <p>The result is a practical prototype for <strong>systemic risk monitoring</strong>, sitting between quantitative finance, complex systems, and AI-assisted decision support.</p> <p>You can explore the project here:</p> <ul> <li><strong> </strong></li> <li><strong> </strong></li> </ul>https://stefano-blando.github.io/en/blog/microsoft-ai-dev-days-risksentinel/Sun, 15 Mar 2026 00:00:00 +0000https://stefano-blando.github.io/en/blog/microsoft-ai-dev-days-risksentinel/<p><strong>RiskSentinel</strong> is the project I built for the <strong>Microsoft AI Dev Days Hackathon 2026</strong>.</p> <p>The core idea was to create a system that could make systemic risk more explorable and more tangible: instead of treating contagion as an abstract output in a paper or a notebook, I wanted an interface where shocks could be launched, propagated, compared, and interpreted in real time.</p> <p>The project combines <strong>network science</strong>, <strong>contagion modeling</strong>, and <strong>agentic AI</strong> on top of research-grade financial network data covering <strong>210 S&amp;P 500 stocks</strong> and <strong>3,081 daily snapshots</strong>. Under the hood, it integrates three propagation models, interactive network analytics with Streamlit and Plotly, and an agentic workflow built with <strong>Microsoft Agent Framework</strong> and <strong>Azure OpenAI</strong>.</p> <p>What I like most about this project is that it sits exactly at the boundary between my research interests and practical prototyping: financial networks, complex systems, decision support, and AI agents all in the same tool.</p> <p>👉 <strong> </strong><br> GitHub: <strong> </strong><br> App: <strong> </strong></p>