<?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 Niladri Roy on Medium]]></title>
        <description><![CDATA[Stories by Niladri Roy on Medium]]></description>
        <link>https://medium.com/@niladriroy0?source=rss-33223d3e1919------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*AkKlcEcA-8j4LONrGLKDKw.jpeg</url>
            <title>Stories by Niladri Roy on Medium</title>
            <link>https://medium.com/@niladriroy0?source=rss-33223d3e1919------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 03:48:02 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@niladriroy0/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[The Invisible Engine: A Systems Thinker’s Guide to Computing Fundamentals]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://niladriroy0.medium.com/the-invisible-engine-a-systems-thinkers-guide-to-computing-fundamentals-a015b55a9789?source=rss-33223d3e1919------2"><img src="https://cdn-images-1.medium.com/max/1024/1*iVf5n-TqOUVsDgvRSuvNjA.jpeg" width="1024"></a></p><p class="medium-feed-snippet">OS secrets: How Context Switching and Cache Misses are secretly killing your latency.</p><p class="medium-feed-link"><a href="https://niladriroy0.medium.com/the-invisible-engine-a-systems-thinkers-guide-to-computing-fundamentals-a015b55a9789?source=rss-33223d3e1919------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://niladriroy0.medium.com/the-invisible-engine-a-systems-thinkers-guide-to-computing-fundamentals-a015b55a9789?source=rss-33223d3e1919------2</link>
            <guid isPermaLink="false">https://medium.com/p/a015b55a9789</guid>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[operating-systems]]></category>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[system-design-interview]]></category>
            <category><![CDATA[software-architecture]]></category>
            <dc:creator><![CDATA[Niladri Roy]]></dc:creator>
            <pubDate>Thu, 18 Dec 2025 20:37:06 GMT</pubDate>
            <atom:updated>2025-12-24T10:28:07.780Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Java Strings: The Architectural Secret That Separates Juniors from Elite Engineers]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://niladriroy0.medium.com/java-strings-the-architectural-secret-that-separates-juniors-from-elite-engineers-4cb9b208ae69?source=rss-33223d3e1919------2"><img src="https://cdn-images-1.medium.com/max/1248/1*IDQq3gIvnhAFLEVTrk4Kjg.png" width="1248"></a></p><p class="medium-feed-snippet">Beyond &#x201C;immutable object&#x201D; the architectural governance system for memory, identity, and intent.</p><p class="medium-feed-link"><a href="https://niladriroy0.medium.com/java-strings-the-architectural-secret-that-separates-juniors-from-elite-engineers-4cb9b208ae69?source=rss-33223d3e1919------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://niladriroy0.medium.com/java-strings-the-architectural-secret-that-separates-juniors-from-elite-engineers-4cb9b208ae69?source=rss-33223d3e1919------2</link>
            <guid isPermaLink="false">https://medium.com/p/4cb9b208ae69</guid>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[jvm]]></category>
            <category><![CDATA[software-engineering]]></category>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[java]]></category>
            <dc:creator><![CDATA[Niladri Roy]]></dc:creator>
            <pubDate>Sat, 13 Dec 2025 16:02:39 GMT</pubDate>
            <atom:updated>2025-12-15T00:47:11.977Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Memory Management in Java: The Invisible Engine Developers Forget to Respect]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://niladriroy0.medium.com/memory-management-in-java-the-invisible-engine-developers-forget-to-respect-7a2784537ddd?source=rss-33223d3e1919------2"><img src="https://cdn-images-1.medium.com/max/1536/1*mIm_WlXMntNukWSHfOyoGg.png" width="1536"></a></p><p class="medium-feed-snippet">Java doesn&#x2019;t run your code. It negotiates with it. That&#x2019;s the part most developers miss.</p><p class="medium-feed-link"><a href="https://niladriroy0.medium.com/memory-management-in-java-the-invisible-engine-developers-forget-to-respect-7a2784537ddd?source=rss-33223d3e1919------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://niladriroy0.medium.com/memory-management-in-java-the-invisible-engine-developers-forget-to-respect-7a2784537ddd?source=rss-33223d3e1919------2</link>
            <guid isPermaLink="false">https://medium.com/p/7a2784537ddd</guid>
            <category><![CDATA[java]]></category>
            <category><![CDATA[jvm]]></category>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[garbage-collection]]></category>
            <category><![CDATA[software-development]]></category>
            <dc:creator><![CDATA[Niladri Roy]]></dc:creator>
            <pubDate>Mon, 08 Dec 2025 10:53:25 GMT</pubDate>
            <atom:updated>2025-12-15T00:47:56.541Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[The Overview Effect. The cosmic’s answer to human’s detachment.]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://niladriroy0.medium.com/the-overview-effect-the-cosmic-answer-to-humans-detachment-6979b2081c97?source=rss-33223d3e1919------2"><img src="https://cdn-images-1.medium.com/max/1024/1*F03b6abVvtEX_N3BQJKHVg.png" width="1024"></a></p><p class="medium-feed-snippet">&#x201C;To love without self is to find the one worth choosing, even in the hush of the infinite.&#x201D;</p><p class="medium-feed-link"><a href="https://niladriroy0.medium.com/the-overview-effect-the-cosmic-answer-to-humans-detachment-6979b2081c97?source=rss-33223d3e1919------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://niladriroy0.medium.com/the-overview-effect-the-cosmic-answer-to-humans-detachment-6979b2081c97?source=rss-33223d3e1919------2</link>
            <guid isPermaLink="false">https://medium.com/p/6979b2081c97</guid>
            <category><![CDATA[science]]></category>
            <category><![CDATA[love]]></category>
            <category><![CDATA[psychology]]></category>
            <category><![CDATA[spirituality]]></category>
            <category><![CDATA[memories]]></category>
            <dc:creator><![CDATA[Niladri Roy]]></dc:creator>
            <pubDate>Sun, 19 Oct 2025 00:23:44 GMT</pubDate>
            <atom:updated>2025-12-20T14:23:18.804Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Edge AI: Why the Future of Intelligence Isn’t in the Cloud]]></title>
            <link>https://niladriroy0.medium.com/edge-ai-why-the-future-of-intelligence-isnt-in-the-cloud-02543b4ff6ad?source=rss-33223d3e1919------2</link>
            <guid isPermaLink="false">https://medium.com/p/02543b4ff6ad</guid>
            <category><![CDATA[science]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[cloud-computing]]></category>
            <category><![CDATA[deep-learning]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Niladri Roy]]></dc:creator>
            <pubDate>Wed, 24 Sep 2025 23:01:46 GMT</pubDate>
            <atom:updated>2025-09-24T23:01:46.654Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*crahxn95iADTxrjI4RWAaQ.png" /></figure><p>What if the smartest technology of the next decade doesn’t live in distant data centers — but right in your pocket, your car, or even your refrigerator?</p><p>For the past 15 years, the cloud has been our symbol of progress. We upload, we stream, we compute — always sending our requests to massive servers somewhere far away, waiting for answers to come back. But here’s the surprising twist: the future of artificial intelligence might not be “up there” at all. Instead, it’s moving closer. Much closer.</p><p>We call it <strong>Edge AI</strong>, and it could completely rewrite how humans and machines interact.</p><h3>The Problem with Always Living in the Cloud</h3><p>Imagine you’re driving an autonomous car. A child suddenly runs across the road. Your vehicle has milliseconds to decide — brake, swerve, or freeze. Now, picture what happens if that decision relies on sending data to a server hundreds of miles away, waiting for the cloud to process it, and then receiving an answer.</p><p>Too late.</p><p>Or think about healthcare. A patient wearing a heart-monitoring device suddenly shows signs of cardiac arrest. If the data has to travel to a distant server before an alert is raised, the delay could be deadly.</p><p>This is where <strong>the cloud reveals its weakness</strong>. While it’s powerful and centralized, it struggles with three things that real-time systems desperately need:</p><ol><li><strong>Latency</strong> — waiting for round trips of data.</li><li><strong>Bandwidth</strong> — limited by how much information networks can carry.</li><li><strong>Privacy</strong> — sensitive personal data moving back and forth across networks.</li></ol><p>Humans don’t have the luxury of delay in critical moments. Neither should AI.</p><h3>What Exactly is Edge AI?</h3><p>At its core, <strong>Edge AI</strong> means moving intelligence closer to where the data is generated — on the “edge” of the network. Instead of relying entirely on cloud servers, devices like smartphones, wearables, drones, or industrial robots process data locally, often in real time.</p><p>Think of it like this:</p><ul><li><strong>Cloud AI</strong> is like calling a friend every time you need an answer. Powerful, but slow if you have to do it often.</li><li><strong>Edge AI</strong> is like carrying that friend around with you — answers are immediate, contextual, and private.</li></ul><p>This is made possible thanks to advances in:</p><ul><li><strong>Tiny but powerful processors (like Google’s Edge TPU, Apple’s Neural Engine, NVIDIA Jetson Nano).</strong></li><li><strong>Optimized AI models (smaller, faster, compressed for low-energy devices).</strong></li><li><strong>Smart connectivity (5G and beyond, where edge and cloud complement each other).</strong></li></ul><h3>Your Smartphone Already Knows</h3><p>Let’s bring it home — literally. If you own a recent iPhone or Android device, you’ve already been using Edge AI without realizing it.</p><ul><li><strong>Face Unlock (Apple Face ID or Android equivalents):</strong> Your face is recognized in milliseconds, without sending an image of you to the cloud. That’s edge processing at work.</li><li><strong>Voice Assistants (Hey Siri, Google Assistant):</strong> Many commands — like setting alarms, turning on flashlight — are now processed locally, reducing lag and preserving privacy.</li><li><strong>Camera Enhancements:</strong> Night mode photography uses AI models directly on your device, turning near-darkness into crisp images in real time.</li></ul><p>Imagine if each of those requests had to ping a distant server. You’d wait seconds instead of fractions of one. The experience wouldn’t feel “magical.”</p><h3>Edge AI in Healthcare</h3><p>Now, let’s step into the hospital.</p><p>Hospitals are increasingly deploying <strong>AI-enabled monitoring systems</strong>. For instance, wearable patches can track heart rhythm, oxygen levels, and blood pressure. Instead of constantly streaming raw data to the cloud, the wearable itself does the first layer of analysis:</p><ul><li>If patterns look normal → data is stored locally, summarized later.</li><li>If irregularities appear → an alert is triggered instantly.</li></ul><p>Companies like <strong>Philips</strong> and <strong>Medtronic</strong> are already piloting such systems. Imagine a patch that notices arrhythmia before the patient does and calls a nurse within seconds. That’s the life-saving edge.</p><h3>Smarter, Safer Cities</h3><p>Edge AI isn’t just personal — it’s urban.</p><p>Think of <strong>traffic management</strong>. Traditional city surveillance cameras send feeds to central servers, where software analyzes them. But with <strong>edge-based cameras</strong>, video can be processed directly inside the camera itself: detecting accidents, recognizing license plates, or optimizing traffic signals on the spot.</p><p>This avoids sending terabytes of video over networks, slashes costs, and ensures cities can respond faster to emergencies.</p><p>Now scale this: tens of thousands of edge cameras making micro-decisions every second. That’s a city brain — distributed, decentralized, alive.</p><h3>Why Now?</h3><p>You might wonder: if this makes so much sense, why hasn’t everything already moved to the edge?</p><p>The answer lies in <strong>hardware evolution</strong>. Ten years ago, AI models were too bulky and processors too weak to run them locally. But today:</p><ul><li>A smartwatch is more powerful than the supercomputers of the 90s.</li><li>AI models like <strong>MobileNet</strong> or <strong>TinyML</strong> shrink algorithms into efficient packages.</li><li><strong>5G networks</strong> make it possible to balance edge + cloud, deciding intelligently what must be done locally vs remotely.</li></ul><p>The timing is perfect.</p><h3>Edge vs Cloud: Not Enemies, but Partners</h3><p>It’s tempting to see this as an “Edge vs Cloud” battle. But the truth is — they’re complementary.</p><ul><li><strong>Edge handles immediacy</strong>: real-time tasks, privacy-sensitive data, and bandwidth-heavy workloads.</li><li><strong>Cloud handles scale</strong>: training massive AI models, long-term storage, and heavy computation.</li></ul><p>Think of it like the human brain:</p><ul><li>Reflexes (edge) happen instantly in your spinal cord.</li><li>Deep thinking (cloud) happens in your prefrontal cortex.</li></ul><p>The combination makes us adaptable, efficient, and intelligent.</p><h3>The Bigger Picture: Human Takeaways</h3><p>So, what does this mean for us — beyond the buzzwords?</p><ol><li><strong>Expect smarter devices, not just smarter servers.</strong> The intelligence of the future will sit in your wristwatch, your fridge, your car, not just “somewhere in the cloud.”</li><li><strong>Privacy will (hopefully) get better.</strong> Because your data doesn’t need to leave your device for everything, you gain more control.</li><li><strong>Speed becomes invisible.</strong> The best technology feels like it disappears — the way your phone unlocks instantly with your face, or how an autonomous car will someday brake faster than you can blink.</li></ol><p>And maybe the most profound shift: we stop thinking of technology as something <em>out there</em> — a distant, untouchable cloud — and start seeing it as something <em>right here</em>, woven into the fabric of our everyday lives.</p><h3>Closing Thought</h3><p>When a heartbeat is irregular, when a car must stop, when a city needs to breathe — waiting for the cloud isn’t enough.</p><p>The future of intelligence isn’t about distance. It’s about presence.</p><p>And that future is already in your hands.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=02543b4ff6ad" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[No Free Lunch Theorem]]></title>
            <link>https://niladriroy0.medium.com/no-free-lunch-theorem-machine-learning-93e5199651ae?source=rss-33223d3e1919------2</link>
            <guid isPermaLink="false">https://medium.com/p/93e5199651ae</guid>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[science]]></category>
            <category><![CDATA[psychology]]></category>
            <dc:creator><![CDATA[Niladri Roy]]></dc:creator>
            <pubDate>Sat, 13 Sep 2025 11:50:45 GMT</pubDate>
            <atom:updated>2025-09-13T14:26:25.929Z</atom:updated>
            <content:encoded><![CDATA[<h3>No Free Lunch Theorem: Why Algorithms Don’t Have a Universal Hack</h3><p>Last winter, I was sitting in a crowded Chennai coffee shop, laptop glowing with half-finished code. On my left, a data scientist was swiping through dating apps. On my right, a trader was bragging about a new algorithm that “beats the market.”</p><p>Here’s the strange part: both screens looked oddly similar. Rows of options, decisions being made, a “best choice” flashing up every few seconds. It made me wonder: <strong>if algorithms can choose dates and stocks… why can’t one algorithm just solve everything?</strong></p><p>That’s when I remembered the unsettling truth: <strong>the No Free Lunch` Theorem.</strong></p><h3>Why We Keep Hunting for the Perfect Hack</h3><p>We humans crave certainty. We want the best recipe, the best workout, the best algorithm. A universal hack. But life doesn’t bend that way.</p><p>I learned this the hard way during my first machine learning project. I tried five algorithms on a messy healthcare dataset — decision trees, SVMs, neural networks, random forests and k-NN. Each one sparkled on some metrics and collapsed on others. I kept thinking, <em>there has to be a single winner</em>.</p><p>Except, there wasn’t.</p><p>The frustration wasn’t just technical. It felt personal, almost existential. I wanted the comfort of a tool that always works. Instead, I found a law of the universe whispering: <strong>there is no universally best solution.</strong></p><h3>The Cosmic Fairness of Algorithms a.k.a. The No Free Lunch Theorem</h3><p>So, what exactly is the <strong>No Free Lunch (NFL) Theorem</strong>?</p><p>In plain words:</p><blockquote><em>If you average the performance of every possible algorithm across every possible problem, they all tie.</em></blockquote><p>No one wins. No one loses. It’s cosmic fairness at work.</p><p>Think of it like cooking. Imagine rating every chef by how they cook every possible ingredient. The one who makes heavenly pasta may burn sushi. The sushi master may struggle with soufflés. On average, over <em>all</em> foods, everyone balances out.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uLvILwYtVnm_hcSafFXMxA.png" /></figure><p>That’s the essence of the theorem.</p><p><strong><em>Algorithms are chefs. Datasets are ingredients. There’s no one kitchen god.</em></strong></p><p>The “aha” moment for me came while explaining this to a non-tech friend. I said:</p><blockquote>“It’s like asking: who’s the best driver? Someone amazing on icy roads, or someone blazing fast on racetracks? The truth is — no one’s best everywhere.”</blockquote><p>She looked up and said, “So… machine learning is just like dating?” “Exactly,” I replied. “Because just like dating, the best match isn’t about perfection everywhere — it’s about finding the one algorithm that works beautifully on your dataset.”</p><p><em>And then I made the mistake of looking into her eyes.</em> <em>I looked into her eyes — and suddenly forgot about datasets, loss functions, everything. They weren’t just eyes; they were the kind of eyes that make you believe algorithms could never capture this — beauty with brains — an unfair advantage in any universe.</em></p><p>Oh, wait… I’m supposed to be talking about the theorem, not about how her gaze felt like a secret variable no model could explain. But that’s the point: we don’t just run algorithms; we get entangled in them.</p><h3>Netflix Knows You… Sort Of</h3><p>Let’s make it concrete.</p><p>When you open Netflix, it doesn’t show you <em>the</em> best show. It shows you <em>your</em> best show — the one its algorithm predicts you’ll like.</p><p>But here’s the catch: the algorithm that works beautifully for recommending comedies may stumble on documentaries. The features that matter for horror movies (suspense pacing, jump scares) aren’t the same as those for stand-up specials (audience laughter peaks).</p><p>Behind the curtain, Netflix runs dozens of models. Some focus on collaborative filtering (“people like you liked this”), others on deep learning embeddings. No single one is the “supreme recommender.” Instead, engineers mix and match, choosing what fits the content.</p><h3>The Science Behind the Limits (Not Yours, Don’t Worry)</h3><p>The NFL theorem rests on a simple but profound symmetry. If an algorithm performs better on some class of problems, then necessarily it performs worse on another class.</p><p>Why? Because “problem space” is unimaginably huge. Every dataset you’ve seen — medical, financial, biological, even dating — is just a speck. Out there in the space of <em>all possible datasets</em>, weird alien problems exist that will break your favorite model.</p><p>In mathematical terms:</p><ul><li>Imagine all functions mapping inputs to outputs.</li><li>Every algorithm is a strategy to guess those functions.</li><li>For every win, there’s an equal and opposite loss elsewhere.</li></ul><p>It’s like karma encoded in computation.</p><h3>Otherwise, if You Want to Explore</h3><ol><li><strong>Healthcare Diagnosis</strong><br>When I helped a team trial ML models for cancer detection, one model shined on lung scans but faltered on liver scans. Doctors wanted “the best.” We had to explain: there’s no universal best. Just the best for <em>this</em> organ, <em>this</em> dataset.</li><li><strong>Stock Market Algorithms</strong><br>I once shadowed a friend at a trading desk. His algorithm printed money during low-volatility weeks but tanked during crises. He shrugged: “There’s no holy grail. You rotate.” That’s NFL embodied.</li><li><strong>Travel Apps</strong><br>Think of Google Maps. Sometimes it optimizes for time, sometimes for traffic, sometimes for scenic routes. If you demand “best in all ways,” you’ll end up stuck.</li></ol><blockquote>Each story drove home the same lesson: the theorem isn’t just abstract math. It’s a mirror of life’s diversity.</blockquote><h3>Why This Matters for Us</h3><p>Here’s the kicker: NFL isn’t telling us to give up. It’s teaching us to specialize.</p><ul><li>In <strong>machine learning</strong>, it means: don’t hunt for the perfect algorithm. Hunt for the perfect algorithm <em>for your problem</em>.</li><li>In <strong>life</strong>, it whispers: stop searching for universal hacks. What works for your friend may fail for you. What heals today may not tomorrow.</li></ul><p>This reframing is oddly liberating. Instead of despairing that no algorithm wins everywhere, we celebrate that algorithms are tools in a toolbox. Pick wisely.</p><h3>The Mindset Shift</h3><p>When I first embraced this, my frustration turned into curiosity. I began approaching datasets like a craftsman choosing chisels: sometimes a blunt but sturdy tool, sometimes a delicate one.</p><p>And beyond tech, I noticed the same principle. The workout that energized me in my teens exhausted me in my twenties. The career advice that lifted a friend left me cold. <strong>The “best” was contextual. Always.</strong></p><p>That realization felt less like a loss, more like freedom.</p><h3>The Future Glimpse</h3><p>Ironically, the NFL theorem may be the reason <a href="https://www.automl.org/automl/"><strong>AutoML</strong></a><strong> </strong>and <a href="https://www.ensemble.com/"><strong>ensemble systems</strong></a> are booming. Instead of pretending one algorithm rules all, we now build systems that adapt — blending models, switching strategies, stacking learners.</p><p>The dream isn’t to escape the theorem. It’s to dance with it.</p><p>Maybe someday, our software won’t promise “one model to rule them all,” but “a thousand tiny models that learn when to shine.”</p><h3>Takeaway</h3><p>The <strong>No Free Lunch Theorem</strong> is less a limitation, more a compass.<br> It reminds us:</p><ul><li>There is no universal best.</li><li>Specialization and context matter.</li><li>Diversity of tools is a strength, not a weakness.</li></ul><p>In machine learning. In cooking. In love. In life.</p><h3>Closing Line</h3><p>As I left that coffee shop — one man still swiping, the other still trading — I smiled at the symmetry. We’re all searching for the universal hack. And maybe the most human truth of all is this:</p><blockquote><strong>there’s no free lunch, but the menu is endless — and the joy isn’t in finding <em>the</em> dish, it’s in choosing <em>yours</em>.</strong></blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=93e5199651ae" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[When Fiction Becomes a Warning: Zola’s Algorithm and the Real-World Science of Predicting Human…]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://niladriroy0.medium.com/when-fiction-becomes-a-warning-zolas-algorithm-and-the-real-world-science-of-predicting-human-f6c382e86f39?source=rss-33223d3e1919------2"><img src="https://cdn-images-1.medium.com/max/1536/1*OxizVpLP8m_Pz7UCZd1Psw.png" width="1536"></a></p><p class="medium-feed-snippet">&#x201C;It predicts not what people do&#x2026; but who they&#x2019;ll become.&#x201D;</p><p class="medium-feed-link"><a href="https://niladriroy0.medium.com/when-fiction-becomes-a-warning-zolas-algorithm-and-the-real-world-science-of-predicting-human-f6c382e86f39?source=rss-33223d3e1919------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://niladriroy0.medium.com/when-fiction-becomes-a-warning-zolas-algorithm-and-the-real-world-science-of-predicting-human-f6c382e86f39?source=rss-33223d3e1919------2</link>
            <guid isPermaLink="false">https://medium.com/p/f6c382e86f39</guid>
            <category><![CDATA[psychology]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[science]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[deep-learning]]></category>
            <dc:creator><![CDATA[Niladri Roy]]></dc:creator>
            <pubDate>Tue, 19 Aug 2025 17:01:59 GMT</pubDate>
            <atom:updated>2025-08-19T17:01:59.499Z</atom:updated>
        </item>
    </channel>
</rss>