<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Dewmini Gunasekera on Medium]]></title>
        <description><![CDATA[Stories by Dewmini Gunasekera on Medium]]></description>
        <link>https://medium.com/@dewminigunasekera?source=rss-562cbe12a2b8------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*GVanC9KHpueAsoyy.jpg</url>
            <title>Stories by Dewmini Gunasekera on Medium</title>
            <link>https://medium.com/@dewminigunasekera?source=rss-562cbe12a2b8------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 04:02:20 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@dewminigunasekera/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Quant Surveillance System — Layered ML — based Insider Trading Detection]]></title>
            <link>https://medium.com/@dewminigunasekera/quant-surveillance-system-layered-ml-based-insider-trading-detection-fe69849b574a?source=rss-562cbe12a2b8------2</link>
            <guid isPermaLink="false">https://medium.com/p/fe69849b574a</guid>
            <category><![CDATA[risk-management]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[quant]]></category>
            <category><![CDATA[market-surveillance]]></category>
            <category><![CDATA[market-risk]]></category>
            <dc:creator><![CDATA[Dewmini Gunasekera]]></dc:creator>
            <pubDate>Fri, 02 May 2025 02:27:04 GMT</pubDate>
            <atom:updated>2025-05-02T02:27:04.068Z</atom:updated>
            <content:encoded><![CDATA[<h3><strong>Quant Surveillance System — Layered ML — based Insider Trading Detection</strong></h3><p>I independently built a full-stack quant surveillance system capable of detecting insider and manipulative trading behavior.</p><p>This system combines domain-simulated trade behavior with anomaly detection, interpretability, and temporal intelligence — built for capital markets, not generic fraud detection.</p><p>It only took 100 mins for me to build this system.</p><h3>Why I Built It</h3><p>Most anomaly models ignore tested wisdom if rigor is embedded and vice versa. I wanted to simulate realistic behavior (using traditional insights), then, flag anomalies statistically, then model behavioral sequences using ML— all in one pipeline.</p><p>This wasn’t coursework. This wasn’t a portfolio filler. This was execution-driven quant design.</p><h3>Architecture — 6 Layers</h3><ol><li><strong>Behavioral Injection</strong> — Simulated large trades, incremental layering, and end-of-window price manipulation</li><li><strong>Feature Engineering</strong> — Encoded timestamp, price, quantity, side, and asset class</li><li><strong>Isolation Forest</strong> — Detected statistical anomalies</li><li><strong>Decision Tree</strong> — Made outputs interpretable for human reviewers</li><li><strong>Flag + Explain Layer</strong> — Created explainable triggers per trade</li><li><strong>LSTM</strong> — Modeled trader-level temporal behavior sequences</li></ol><h3><strong>Results</strong></h3><p>A consistent Precision of 1.00 over multiple sets generated.</p><p>A recall rate around 0.5 over multiple sets.</p><p>This means that there are no false positives and tends to pick half of all insider traders. But when traders perform the insider trades over considerable time, their profiles may be detected.</p><h3>Project Link</h3><p><strong>GitHub:</strong> <a href="https://github.com/dewminigunasekera/quant-surveillance-system">Quant Surveillance System</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fe69849b574a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[“Riders of Justice”: A Study of Genius, Grief, and Charisma Rewritten]]></title>
            <link>https://medium.com/@dewminigunasekera/riders-of-justice-a-study-of-genius-grief-and-charisma-rewritten-2799e59d2823?source=rss-562cbe12a2b8------2</link>
            <guid isPermaLink="false">https://medium.com/p/2799e59d2823</guid>
            <category><![CDATA[storytelling]]></category>
            <category><![CDATA[film-analysis]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[hacking]]></category>
            <category><![CDATA[character-studies]]></category>
            <dc:creator><![CDATA[Dewmini Gunasekera]]></dc:creator>
            <pubDate>Thu, 01 May 2025 14:18:30 GMT</pubDate>
            <atom:updated>2025-05-01T14:18:30.822Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*n6H1x53z2MicjCq21Lb4sA.jpeg" /><figcaption><em>Still from</em> Riders of Justice <em>(2020). A film where violence unravels into introspection, and genius hides behind broken men who never asked to be seen. courtesy of Magnolia Pictures</em></figcaption></figure><blockquote>A breakdown of Riders of Justice — where a Bayesian monk, an elite hacker, and a secluded ML engineer face randomness, grief, and a missing pillar of power: charisma.</blockquote><p><strong>Introduction</strong></p><p><em>Riders of Justice</em> appears, at first glance, to be a revenge thriller. But beneath the gunfire lies a quietly explosive narrative — one that explores the wounded psyche, the neglected genius, and the terrifying beauty of randomness. This film doesn’t just subvert genre — it redefines it, asking not for violence, but for peace. Not generic peace — but peace in your own form.</p><p>It also shows the <strong>true data scientist</strong>, the <strong>great storyteller</strong>, the <strong>true hacker</strong>, the <strong>elite ML engineer living in seclusion</strong>, and the <strong>professional soldier</strong> whose rationality is built on discipline and moral clarity. Each one is a humane, wounded, and rational version of what society never learns to embrace until chaos forces their value into the open.</p><p><strong>1. The Elite Scientists: Geniuses in Exile</strong></p><p><strong>Otto — The Composed Bayesian Mind, Crushed by Random Karma</strong></p><p>Otto is not erratic or broken. He is composed, high-minded, and quietly dignified — a true <em>Bayesian monk</em> of the rational world. His suffering is karmic — random, undeserved, and sudden. And rather than collapsing into rage or delusion, he seeks to understand. Otto doesn’t break — he endures, calculates, and <em>opens</em>. His journey is not one of healing from madness, but of <strong>finding meaning in misfortune,</strong></p><p>In karmic thinking, not all suffering is due to internal faults. Sometimes, we face <em>residual karma</em> — effects of old causes not under our current control. Otto’s path is one of <strong>acceptance, logic, and quiet service</strong>. He represents <strong>the noble mind dealing with unjust outcomes</strong> without losing grace.</p><p>Otto is the film’s <strong>true data scientist</strong>, applying statistical logic to human suffering. And more than that — he is the <strong>great storyteller</strong>, weaving narrative coherence out of randomness and giving meaning to the chaos around them.</p><p><strong>Lennart — The Elite Hacker</strong></p><p>A brilliant surveillance and intrusion expert, plagued by abuse and trauma. Not just a coder — but a system-level thinker with pain underneath every exploit. His genius is undeniable, but his mind is turbulent. He navigates the digital world better than the emotional one.</p><p>Lennart is the <strong>true hacker</strong>: unpredictable, precise, emotionally scarred, yet indispensable in understanding and confronting the system’s vulnerabilities — both technological and human.</p><p><strong>Emmenthaler — The ML Engineer Living in Seclusion</strong></p><p>A deeply technical genius with social dysfunction, building models while managing trauma. Misunderstood, unstable, but precise — a prototype of the <em>unseen ML mind</em> in modern society. Tinkers with algorithms and simulations but is never invited to the table of “real” decision-making.</p><p>Emmenthaler is a symbol of the elite ML engineer who builds powerful systems from the shadows — a recluse, likely by choice, and financial hardship. Unlike Otto and Lennart — who are wounded but materially stable — Emmenthaler’s isolation is compounded by poverty.</p><blockquote><em>These men are each </em><strong><em>lesser egoist and more humane versions of Walter White from Breaking Bad</em></strong><em> — elite, invisible, and dangerous only when finally cornered.</em></blockquote><p><strong>2. Markus: The Rational Avenger</strong></p><p>A soldier and father whose <strong>instinct is order through force</strong>. His logic is binary: A caused B, therefore destroy A. He is not stupid — he is disciplined, determined, and emotionally locked. His trauma silences him. His silence isolates him. His isolation becomes destruction.</p><p>Markus is the <strong>rational professional</strong>. His skillset comes from lived experience, physical training, and moral conviction. He is the <strong>executor</strong> of justice — driven by precision, not chaos. But even he must learn: not all violence clarifies, and not all pain can be controlled.</p><p>He is the <strong>professional soldier</strong>, and through him, the film honors a form of order that is often denied its depth: rational, strategic, loyal — and ultimately vulnerable.</p><p><strong>3. The Two Core Archetypes</strong></p><p><strong>Mind (Scientists)</strong></p><ul><li>Control through logic, intellect, and systems.</li><li>Emotionally repressed, socially misaligned.</li><li>Their war is internal: <em>can I model the chaos and still be human?</em></li></ul><p><strong>Action (Markus)</strong></p><ul><li>Control through movement, certainty, and protection.</li><li>Feels deeply, but expresses through precision violence.</li><li>His war is external: <em>can I eliminate the threat and still be a father?</em></li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*BTjYLwFfG7gzcTn_p2blYw.jpeg" /><figcaption>The unlikely cohesion of Protagonists in form against the Antagonist. <em>courtesy of Magnolia Pictures</em></figcaption></figure><p><strong>4. The Missing Pillar: Charisma and the Antagonist</strong></p><p>While the protagonists embody logic and discipline, they lack one defining force: <strong>charisma</strong>. That elusive third pillar — emotional magnetism, expressive confidence, and presence — is embodied not by the heroes, but by <strong>Kurt</strong>, the gang leader.</p><p>Kurt radiates confidence, control, and style. He is composed but expressive, commanding every room he enters. He is dangerous, yes — but <em>he is also memorable</em>. His words, look, and stance carry authority. Unlike Markus and the scientists, <strong>he doesn’t need to be right to be followed — he simply needs to be seen.</strong></p><p>This reveals an unsettling truth:</p><blockquote>The world sometimes follows those with charisma, not those with insight or moral clarity.</blockquote><p>The protagonists struggle with isolation. Kurt does not. He draws people into orbit — even in evil. The scientists compute. Markus acts. But <strong>Kurt influences</strong>.</p><p>This is the unspoken triangle of human forms in <em>Riders of Justice</em>:</p><ol><li><strong>Mind</strong> — represented by Otto and the others</li><li><strong>Action</strong> — represented by Markus</li><li><strong>Charisma</strong> — embodied by the villain</li></ol><p>And that missing pillar makes the film honest: It doesn’t give the good guys every tool. It makes them win <strong>without charisma</strong>, and that is its most subversive triumph.</p><p><strong>5. What the Film Portrays : The Discipline of Openness</strong></p><ul><li>The scientists must <strong>listen to society and the practicality</strong>, not just interpret it through internal value code.</li><li>The executors must <strong>listen to family and the smaller circle</strong>, not just guard it.</li><li>Each character learns that <strong>healing doesn’t mean changing your form</strong> — it means <strong>opening your form to others</strong>.</li></ul><p><em>It’s not enough to be right. You must be present.</em></p><p><strong>6. The True Climax: A Dinner Table, Not a Gunfight</strong></p><p>The Christmas scene is the real victory.<br> A statistician, a soldier, a hacker, and an AI engineer — sharing a meal.<br> All of them still who they are, but finally <strong>seen</strong>, <strong>heard</strong>, and <strong>at peace</strong>.</p><p><strong>Conclusion: Peace, In Your Own Form</strong></p><p><em>Riders of Justice</em> is not about revenge.<br> It’s about <strong>personal healing</strong>, <strong>scientific isolation</strong>, and the need for <strong>human re-entry</strong>.<br> It’s about finding <strong>meaning in grief</strong>, not by becoming someone else, but by <strong>being accepted as you are</strong>.</p><blockquote><em>You don’t need to soften your mind or your body.<br> You just need to open it — to grief, to randomness, and to those who care.</em></blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2799e59d2823" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>