<?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 Kevin Xu on Medium]]></title>
        <description><![CDATA[Stories by Kevin Xu on Medium]]></description>
        <link>https://medium.com/@xukevinwork?source=rss-711bee898a2c------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*WNlOAnLTHACtVC9V.jpg</url>
            <title>Stories by Kevin Xu on Medium</title>
            <link>https://medium.com/@xukevinwork?source=rss-711bee898a2c------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 18:22:54 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@xukevinwork/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[Punjab, Dissemination, Agricultural Knowledge]]></title>
            <link>https://xukevinwork.medium.com/punjab-dissemination-agricultural-knowledge-4e77bf34134b?source=rss-711bee898a2c------2</link>
            <guid isPermaLink="false">https://medium.com/p/4e77bf34134b</guid>
            <dc:creator><![CDATA[Kevin Xu]]></dc:creator>
            <pubDate>Tue, 01 Feb 2022 23:05:36 GMT</pubDate>
            <atom:updated>2022-02-01T23:05:36.355Z</atom:updated>
            <content:encoded><![CDATA[<p>I spend some time working with my university’s policy center to do research about climate mitigation-related policy.</p><p>In this article, I go over a policy recommendation I came up with to target the core problem of</p><blockquote>Increasing Rates of Agriculture Tech Adoption among Farmers in Punjab, India</blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4e77bf34134b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Climate, Chad, and Humanitarian Crisis]]></title>
            <link>https://xukevinwork.medium.com/climate-chad-and-humanitarian-crisis-cea8f452171a?source=rss-711bee898a2c------2</link>
            <guid isPermaLink="false">https://medium.com/p/cea8f452171a</guid>
            <dc:creator><![CDATA[Kevin Xu]]></dc:creator>
            <pubDate>Tue, 01 Feb 2022 22:38:20 GMT</pubDate>
            <atom:updated>2022-02-01T23:05:08.800Z</atom:updated>
            <content:encoded><![CDATA[<p>I spend some time working with my university’s policy center to do research about climate mitigation policy.</p><p>The past semester, I’ve been looking into climate-related problems within the Lake Chad basin with other Penn students.</p><h3>Background — Why Chad</h3><p>Why did I choose Chad?</p><p>Lake Chad is one of the largest lakes in Africa and supports the livelihoods of over 30 million people (<a href="https://www.wilsoncenter.org/sites/default/files/media/uploads/documents/Climate%20Change%20and%20Violent%20Extremism%20in%20the%20Lake%20Chad%20Basin%20Key%20Issues%20and%20Way%20Forward_0.pdf">Frimpong, 2020</a>). However, over the last six decades, climate variability has significantly depleted the size of the lake and its resources. In the 1960’s, Lake Chad had an area of over 26,000 square kilometers but has dwindled to less than 1,350 square kilometers by 2014 (<a href="https://www.wilsoncenter.org/sites/default/files/media/uploads/documents/Climate%20Change%20and%20Violent%20Extremism%20in%20the%20Lake%20Chad%20Basin%20Key%20Issues%20and%20Way%20Forward_0.pdf">Frimpong, 2020</a>). This represents a 90% decrease in water volume and a 90% decrease in aquatic commodity production (<a href="https://www.wilsoncenter.org/sites/default/files/media/uploads/documents/Climate%20Change%20and%20Violent%20Extremism%20in%20the%20Lake%20Chad%20Basin%20Key%20Issues%20and%20Way%20Forward_0.pdf">Frimpong, 2020</a>). The alarming depletion of the lake’s resources is a result of increased temperatures, desertification, soil infertility, and other environmental degradations. An estimated 2.3 million refugees are displaced within the Lake Chad region due to issues relating to water and over 10.7 million people are impacted by the lake degradation<a href="https://www.climate-refugees.org/spotlight/tag/Lake+Chad"> (Tower, 2020)</a>.</p><p>The problems that exist in Lake Chad are also not entirely unique to the rest of the world. For example, Lake Urmia and the Aral Sea in Central Asia have seen similar drastic reductions in water volume in recent years due to climate change. The Kazakhstan basin faces potential similar problems of displacement and violence. (<a href="https://www.mdpi.com/2073-4441/12/3/749/pdf">Tussupova, 2019</a>). The Lake Chad situation is particularly dire, and it’s something I was interested in. What’s happening in Chad is also relevant for the rest of the world</p><h3>The Findings</h3><p>I’ve summarized some of the findings below regarding problems, and added some excerpts in quotes from research I’ve done. We can summarize the problems into two main buckets</p><h3>Legal Ambiguity</h3><p>Pre-existing lack of consensus in the international community on terminology such as climate migrants has rendered many interventions ineffective or otherwise complicated.</p><h3>Exacerbating Existing Conflict</h3><p>Climate change has exacerbated existing obstacles for the inhabitants seeking relocation and created new ones. In addition to issues such as violence, poverty, and food insecurity</p><h3>Legal Issues: A Deeper Dive</h3><p>There’s significant ambiguity on whether people affected by climate change qualify as refugees under the existing legal definitions or if a new framework is needed.</p><p>For example, if climate change induced crop failure leads to economic issues and widespread civil unrest, which factor ultimately pushed people away from their homes? Another important question surrounds issues of socioeconomic status and coping with adverse conditions. A region devastated by extreme heat might not technically be uninhabitable, but only those with sufficient financial resources would be able to maintain their livelihoods in the increasingly adverse conditions. How should the international community refer to displaced individuals in such circumstances, are they economic or climate migrants? These nuances in definitions impact the policies surrounding this issue and are a critical consideration at the foundation of this policy project.</p><p>Equally important in contributing to the ambiguity surrounding climate change-driven migration is the distinction between migrants and refugees. Migrants are people who have voluntarily moved, either within their country of origin or across borders. The factors leading migrants to move may be multifaceted but at the highest level, the most widely accepted definition of migrants is that they can potentially return home without fearing for their safety. Some factors that may cause people to migrate are natural disasters, famine, extreme poverty, and education. In contrast, refugees require ‘international protection’ after moving outside their country of origin out of fear of, among other things, persecution, conflict, or violence. Refugees require sanctuary in countries outside of their home countries because returning home would be too dangerous for any number of reasons. While the terms migrants and refugees are often conflated, the legal protections defined for these two groups under existing international law vary (<a href="https://www.unhcr.org/en-us/news/latest/2016/3/56e95c676/refugees-migrants-frequently-asked-questions-faqs.html">UNHCR, 2016</a>).</p><h3>Exacerbating Existing Problems</h3><p>Whereas the legal framework for migrants versus refugees is unclear, the effect of climate on existing problems in the Chad Region is clear.</p><h3>Heat</h3><p>Extreme heat conditions have led to diminished yields, increased transportation times, and conflicts related to geographic shifts, all of which has contributed to forced displacement.</p><p>Historically, heat has been a leading cause of displacement in other regions. 10% of the increase in total migration between 1970 and 2000 is linked to access to water (<a href="https://www.worldbank.org/en/news/press-release/2021/08/23/lack-of-water-linked-to-10-percent-of-the-rise-in-global-migration">World Bank, 2021</a>). Extreme heat events play a critical role in the worsening of droughts, alteration of weather patterns, and creation of conditions vulnerable to wildfires. All of these factors affect access to water. And as the average global temperature rises, once sporadic heat waves are predicted to become longer, more frequent, and more extreme.</p><p>When we look at Chad specifically, one case study on a group of villages on the Chad portion of Lake Chad found that more than half of all households noted a high dependency on seasonal variations in natural assets compared to household assets (Okpara, Stringer &amp; Dougill 2016). This means that the lake tends to be more accessible as a water source for farmers and fishermen than to pastoralists who often must walk long distances around 50km to reach the Lake’s water (Sarch, 2002). The receding lake has created a variety of wetlands, floodplains and swamps which provide fertile soil for food (<a href="https://www.wilsoncenter.org/sites/default/files/media/uploads/documents/Climate%20Change%20and%20Violent%20Extremism%20in%20the%20Lake%20Chad%20Basin%20Key%20Issues%20and%20Way%20Forward_0.pdf">Frimpong, 2020</a>). Unfortunately, farming requires large investments in gasoline for irrigation pumps and digging wells in addition to the cost of fertilizer, pesticides which limits ability to enable larger crop yields (Okpara, Stringer &amp; Dougill 2016).</p><p>Nearly all the participants from Lake Chad in the interviewed expressed concern over having enough money for basic household needs such as food, water, clothing, and housing (Okpara, Stringer &amp; Dougill 2016). Few households have access to credit and market opportunities for harvested produce is limited due to poor rural road systems (Okpara, Stringer &amp; Dougill 2016). Although fishing activities themselves were not thought to have decreased amongst the inhabitants, the size and quantity of fish catches have declined while the distance travelled, and costs associated with fishing have grown tremendously. This can be seen in a FAO report stating a decrease from 220,000 tonnes of fish in the 1960’s to about 100,000 tonnes in 2000 and further declines since (Okpara, Stringer &amp; Dougill 2016; FAO, 2012).</p><p>Overall, the shrinking of Lake Chad has had several impacts on the livelihood of its inhabitants though they are often difficult to directly quantify but can be seen through a reduction of household livelihood and resiliency as more individual families attempt to diversify their income streams in an attempt to cope with receding waters.</p><h3>Conflict</h3><p>Chad has a long-standing, complex web of historical violence and conflict, and insurgency groups and warring actors continue to target the increasingly large vulnerable population to further their agenda.</p><p>Insurgency groups continue to attack herders, and seize herds of cattle to finance their operations or feed on in their hideouts (Takway, 2020). Insurgency groups have accelerated the formation of regional alliances with pastoralists and farmers, promising an exchange of convenience.</p><p>More recently, the area has become a stronghold for “a large number of former fighters of the Chadian and Central African civil wars and the Boko Haram insurgency and other groups, such as Al Qaeda in the Islamic Maghreb (AQIM), Daesh or the Islamic State (IS) and multiple splinter or allied groups in the greater Sahel-Sahara, ranging from Mauritania, Mali, Algeria” (<a href="https://www1.undp.org/content/dam/oslo-centre/documents/Zebulon%20Suifon%20Takwa_v3%20(1).pdf">Takway, 2020</a>). Between May 2011 and 31 July 2020, more than 37,500 people were killed in the conflict (<a href="https://www.downtoearth.org.in/blog/climate-change/climate-change-conflict-what-is-fuelling-the-lake-chad-crisis-75639">Mohanty, 2021</a>). Though military forces fight against these armed groups — focusing on a strategy of military containment — they have barely been effective in preventing civilians from combat and terrorism. In the Lake Chad Region as a whole, around 17 million people are living in areas affected by violence resulting from the conflict between official state militaries and non-state armed groups (<a href="https://www.doctorswithoutborders.org/lake-chad-crisis">Doctors Without Borders</a>).</p><p>Climate change has directly made this conflict worse in some aspects. As the former lakebed dries up, for example, small islands emerging from formerly submerged environments have become hideouts and chokeholds for insurgents (Boko Haram and ISWAP, in particular) and criminal groups living with fishing communities. As farmers and pastoralists continue to be displaced, the convenience of alliances with stakeholders in the conflict is increasingly enticing.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/976/1*01kGf4sYz6jxJJ0p5wNYFw.jpeg" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=cea8f452171a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[My Working List of Cool Startup Ideas]]></title>
            <link>https://xukevinwork.medium.com/my-working-list-of-cool-startup-ideas-5809e35b5f4b?source=rss-711bee898a2c------2</link>
            <guid isPermaLink="false">https://medium.com/p/5809e35b5f4b</guid>
            <dc:creator><![CDATA[Kevin Xu]]></dc:creator>
            <pubDate>Sat, 07 Nov 2020 04:56:14 GMT</pubDate>
            <atom:updated>2022-02-23T19:04:04.181Z</atom:updated>
            <content:encoded><![CDATA[<h3>Some Project Ideas</h3><p>Below are some ideas of projects or companies I think would be cool, but I personally don’t care enough about to implement. My long-term interests lie in agri-technology and being one of the people to make long-standing progress toward climate change.</p><h3><strong>Marketing</strong></h3><p>Option 1: Spend $1,000,000 to hire Kylie Jenner to sponsor your brand.</p><p>What you get:</p><ul><li>30 million impressions from Kylie’s followers</li></ul><p>Option 2: Spend $1,000,000 to hire 10,000 mid-sized influencers to promote your brand</p><p>What you get:</p><ul><li>10,000 influencers each with 10,000 followers= 100,000,000 million impressions</li><li>Brand loyalty and more tailored-posts from smaller influencers who care more about their products</li></ul><p>What if we could help a new business find their “option 2” in minutes? The business puts in its details and integrates at the time of account creation. Everything else is one click.</p><h3><strong>Anomaly Detection</strong></h3><p>Problem 1: Agriculture</p><p>Yields are down in a large portion of the farm. We suspect pest breakout or soil degradation began in a few acres of land and spread. Could we have found this in real-time before it spread?</p><p>Problem 2: Cybersecurity</p><p>A virus spread into the network. We suspect a device was a single point of failure. Could we have found this in real-time through analyzing abnormal data?</p><p>Problem 3: Infrastructure</p><p>A water pipeline bursts and we suspect even before then, millions of gallons of water leaked for years. Could we have discovered this sooner?</p><p>These problems fit a general model where a system that was generating consistent data had an anomaly that turned out to be a problem. What if we could deploy a tool across industries that found anomalies in data and let humans decide what to do with those anomalies?</p><p>What if this were an API?</p><p>Palantir helps centralize and standardize data across industries. We could plug this in as a next step for firms that use Palantir’s product. Or firms with their own data streams could run this via the cloud.</p><h3><strong>OS for Small Businesses</strong></h3><p>Imagine you’re a small business.</p><p>What if there was one platform where you could do all the little things needed to run a business: payments, content management, payroll.</p><p>What if you had a CFO, CMO all in one platform</p><p>Think about it like this</p><ol><li>You begin onboarding onto our product</li><li>You enter our existing integrations marketplace. And you get two choices</li></ol><p>A. Connect anything you already use to the platform : Ramp, Mint, Workday and manage it in one place.</p><ul><li>Add on custom extensions to these already great products.</li></ul><blockquote>For restaurants, link your Ramp purchase history to automatic inventory management and accounting. If our system sees a customer bought spaghetti through Ramp, we’ll automatically budget and account for 2 tomatoes and a serving of pasta.</blockquote><p>B. Don’t have these tools. Find the best tool for your need in an instant: Don’t have a CRM platform. We’re partnered with Salesforce and will set you up in one click.</p><h3><strong>News for you</strong></h3><p>Step 1: You install a chrome extension</p><p>Step 2: Search for news on google</p><p>Step 3: The extension offers a “new” news tab, split into three sections</p><ul><li>“Favored” — your favorite source’s results.</li><li>“Left” — A collection of news articles related to your query from left-leaning sources</li><li>“Right” — A collection of news articles related to your query from right-leaning sources</li></ul><p>News <strong>for you</strong></p><h3>An App for Gig Workers</h3><p>There are hundreds of gig economy apps out there, and workers frequently juggle between multiple apps, each at irregular intervals, to sustain their livelihoods.</p><p>Let’s build an app that 1) maximizes profit for workers across apps and 2) minimizes idle time in tasks.</p><p>The app is not focused on finding work, but managing the platforms each worker already uses.</p><h3>(Stale) Idea 4: Debugging</h3><p>Disclaimer: You can do all of this with Loom, so I’ve deleted it. I’m a fan of loom.</p><ul><li>Click 1 — Click, drag, and record your screen to reproduce a bug behavior.</li><li>Click 2 — Provide access to chrome debug logs, and fetch the error messages</li><li>Click 3 — Send the link in an instant to a coworker</li></ul><p>There’s no need to call a client about what the bug is anymore. You have all the information you need. Less meetings, less communication, more productivity!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5809e35b5f4b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Game-ifying High School Debate]]></title>
            <link>https://xukevinwork.medium.com/game-ifying-high-school-debate-3782f80565d?source=rss-711bee898a2c------2</link>
            <guid isPermaLink="false">https://medium.com/p/3782f80565d</guid>
            <category><![CDATA[debate]]></category>
            <dc:creator><![CDATA[Kevin Xu]]></dc:creator>
            <pubDate>Wed, 14 Oct 2020 05:58:51 GMT</pubDate>
            <atom:updated>2022-04-04T05:01:02.053Z</atom:updated>
            <content:encoded><![CDATA[<h3>Debate: A Game</h3><p>I used to do debate in high school, and I did pretty well all things considered (ranked top in the nation for a few years).</p><p>But compared to most of my peers of that caliber, I didn’t spend very much time on the activity or effort in the community.</p><p>I was never the “best” debater.</p><p>I was a poor public speaker for much of high school. While others could talk up a storm, people were surprised to see that I was the type of person to never ask a question in a lecture. I would never say no to a pamphlet handed to me on a street. I would always say sorry when others bumped into me.</p><p>I was also not the hardest worker for much of high school. While others spent days and countless hours researching and preparing for the activity, I watched a lot of youtube about random things I thought were interesting, like supermarkets as a propaganda tool for civil wars.</p><p>In general, I had low stamina. Some of the top debaters I knew were also not the best speakers or the hardest workers, but damn, they could cram before a tournament. I got tired easily and needed my 8 hours of sleep, a poor formula for an activity that runs on weekends from 6 am to 11 pm.</p><p>So why did I do well? <strong>I believed that unequal players playing a game with equal rules lead to unequal outcomes. </strong>I was unequal to most players, so I decided why should I play this debate “game” by the equal rules set in the rulebook. Let me basically change the pace of the game, so that they can play my game, one that I’m pretty good at. (Note: sometimes it was unavoidable to play the game of debate as it was meant to be. In those cases, I tried my best and did okay. I did exceptionally well when the game was the one I designed, and by doing okay in a few and exceptionally well in most, overall I did pretty well).</p><p>What did my version of debate look like versus others? To me, I stripped debate down to its core: Debate is about <strong>convincing</strong> someone you win. The entire activity and the norms that exist are built around that. And there are multiple ways to convince a person, but the overwhelming norm in the community is through meaningful discussions, authentic argumentation, and clash. Why? Maybe that’s the most fruitful way to get something out of the activity. But I used to think to myself: <em>if all debate is convincing then man,</em> <em>there have to be easier ways to do that.</em></p><p><strong>Anyways, here were some strategies that I employed. There are plenty more, but I’ve selected a few.</strong></p><p><strong>Non-Content:</strong></p><p>I used to think that if you make the judge love you, the judge will vote for you. My parents used to tell me, I’d vote for you 100% of the time no matter if I thought you were actually winning. I thought, dang, can I try to be so likable that someone‘s implicit bias is so strong they have no choice but to vote for me.</p><p>For context, I debated on a circuit with a lot of first-generation immigrant judges.</p><p>I figured that if I had trouble understanding some of the arguments we talked about, these parents whose second or third language was English would probably have way more trouble following along with the 4 minutes, 900-word rapid-fire speeches about monetary policy from high schoolers who don’t even understand the policy themselves.</p><p><strong>At first, I thought what’s the point of content? If I tune out for 10 seconds, I’m lost. I’ll wager that most of these judges are tuning out for more than 10 seconds in a 1-hour match and are even more lost.</strong></p><p>If these judges are tuned out, how are they going to make the decision? By voting for who they think “deserves” to win. If they don’t know who “deserves” to win because they got lost, maybe they’ll vote for who they like the most or looks the sharpest in addition to who perceptually seemed like they were winning.</p><p>So instead of content, let me focus on fixing perception. Let me look like the smartest guy in the room. Let me have that nice, sharp tie that others may not have. Let me look as nonchalant as possible while others shake their legs with anxiety before speeches. Let me have a face of chagrin as I hear an opponent’s speech that doesn’t seem to make sense, while others look at my speech with a face that looks like agreement in comparison. Let me focus on being that guy who you go “eh, I have no idea who’s going on but he looked like he won. He’s someone I want to vote for anyway.” Let me speak confidently, continuously, and repeatably talk about how my opponents are misconstruing, misunderstanding, and missing the whole point of the round.</p><p>If you are a judge and I happen to bump into you outside our round earlier, it’s not a coincidence that I learn your kid is applying to college. And we talk a lot about his dream college and his life story and the ways in which he would be a great fit at any top university. We talk about what your kid doesn’t do well, how I can relate, and how there are ways to work around it. You end up being my judge 15 minutes later and dang.</p><p>These were the things that I did, that made up for the holes in content to “win” a debate round.</p><p><strong>Content — Stories</strong></p><p>But all of this is more or less pandering or an act. And the acting is obvious. A lot of judges are sharp and educated. They see through it. They don’t like it. They want to engage in the content. They have kids that are also debating, and complaining about the judges that fall for pampering or that aren’t listening to their argumentation. They don’t want to be that person.</p><p>Pandering is also non-excludable. Everyone can pander, and once people caught on that pandering worked, they learned to do it better than me. I was not the most handsome fellow, not the sweetest talker. I made the effort, but once others made the effort, they were naturally much better than me. I was not the most wealthy. Others soon got nicer suits and made even more outrageous faces. So at a certain point, those non-content strategies became a tool in the bag for the right audience and opponent. But they were far from the silver bullet solution to win all rounds that they once were. At its core, I realized debate was about content, and I couldn’t avoid the content that made the activity meaningful.</p><p>So I went back to content. Why did conveying content have to involve heated discussions about complex topics with a slew of words at an expert-level intensity of discourse, all in the span of a few minutes? If I saw two professors in foreign fields speaking as fast as they could about an issue, I wouldn’t get a thing. That would sound like a foreign language to me. My professors had always used analogies and stories to explain things concisely and in abstractions to me. When they wanted me to relate, they brought up stories and examples of people I could see myself in to visualize and internalize the concepts.</p><p>Why couldn’t I do the same? Why not avoid the trap to engage in a furious debate, and remember debate isn’t something between me and them. It’s about me explaining a concept to a judge and once they’re on board convincing them after.</p><p>Thus, when it came down to crunch time, while others dove into why the policies I advocated for were problematic in the state legislature or law in their two minutes of precious speech time, I used analogies and stories with characters I thought the judge could relate to.</p><blockquote>Like the story of Sam, who was forced to move from his home and drop out from school because housing prices got so expensive he had to leave? Sam’s not an isolated incident. There are 50,000 other Sam’s in the world that exists when you vote for them. It’s that simple. And yes, my world’s not perfect. We need extra money and capital to push legislation through, but it’s for keeping those 50,000 Sam’s in their homes.</blockquote><p><strong>Content- Framing and Confusion</strong></p><p>There were many times in my debate career when I was thoroughly outclassed and outmatched by the other team in preparation and speaking ability.</p><p>I knew in a traditional game, I had lost, but why did it have to be traditional? If we think of debate as a tug of war, why did this have to be a match where if the opponents crossed the middle point by an inch they won? Could I make it such that they only won if they dragged me 50 feet past the middle point? All I had to do was convince the referee that’s how the game was played.</p><p>That’s exactly what I did. Convince the judge for the opponents to win, they had to do much, much more than was necessary. So even if I was losing by a lot, I was losing by enough to lose.</p><p>There were a few ways I did this:</p><p>One was getting my opponents to agree with small concessions about what we had to prove and what they had to prove to win. To them, it seemed like an innocent act: clarifying the round for all stakeholders. But in reality, it was like a trojan horse. Those small concessions became the large parts of my speeches where I trapped them into not meeting the absurd obligations they agreed to.</p><p>One common, interesting strategy was a confusion framework. It went something like</p><blockquote>Our opponents were tasked to prove to you why we should do something. If you’re confused, they haven’t done their job. But I have.</blockquote><p>I would follow that by then confusing the heck out of everyone. A person can only remember 3–4 facts at any given time. If you give them too much information, they’ll forget everything. Usually, the last thing you want to do when convincing someone to vote for you is to get them to forget everything. But if things are terrible, maybe that’s not such a bad thing.</p><p>And that’s what I did. I confused judges as much as I could. I said nonsense, contradicted my own claims, agreed with their claims, denied my own claims, and agreed with the disagreement of my own claims. At the end of the day, when the judge was confused beyond belief, I would wrap it up by bringing it all into our framing.</p><p><strong>Example 4 — Anticipation</strong></p><p>When a debate topic for a given month gets released, you kind of know what people were probably going to argue about on each side. For example, if the topic was about globalization, you’d get arguments like</p><blockquote>Globalization lowers prices for consumers! Globalization causes jobs to leave the country!</blockquote><p>If you knew what people are going to argue, you can definitely throw them off.</p><p>How? Take a look at this snippet from a real round.</p><blockquote>Them: Globalization lowers interest rates! (Implicitly, they and almost everyone assumes that lower interest rates is a good thing. We hear about this all the time in the stock market and in the news).</blockquote><blockquote>Me: Why is that a good thing? I’ll argue that lower interest rates are bad.<br>Me: Here are 3 reasons why lower interest rates causes higher utility prices, more student loans, and war in the middle east (absurd reasons that probably aren’t true, but I’ve spent my time finding evidence for)</blockquote><blockquote>Them: What?</blockquote><blockquote>Me: I have the evidence. Take a look! (I pieced together studies that would “support” what I said. One author who said this was a possibility. Another who found a statistical correlation between the two, etc)</blockquote><blockquote>Them: Uh..</blockquote><p>I basically did this every month with every topic.</p><p>I was a menace.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3782f80565d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Automating Behavioral Interviews]]></title>
            <link>https://xukevinwork.medium.com/automating-behavioral-interviews-987954951f8a?source=rss-711bee898a2c------2</link>
            <guid isPermaLink="false">https://medium.com/p/987954951f8a</guid>
            <dc:creator><![CDATA[Kevin Xu]]></dc:creator>
            <pubDate>Fri, 14 Aug 2020 22:12:27 GMT</pubDate>
            <atom:updated>2022-01-31T03:35:58.554Z</atom:updated>
            <content:encoded><![CDATA[<p>If you’ve ever played League of Legends, this type of interaction is abnormally common:</p><blockquote>usernameponey123: you **fk** suck**</blockquote><blockquote>teammate1: no you’re trashlk. Uninsta**@!!!!!</blockquote><blockquote><em>usernameponey123 has left the game</em></blockquote><p>I used to play, and used to think</p><p><em>I wonder what usernameponeuy123 and teammate1 are like in real life? Are their actions in this game a reflection of their personalities outside of the game?</em></p><p><strong>Can we use games to reveal hidden personality traits?</strong></p><p><strong>What’s the Idea?</strong></p><p>Let’s build an online, competitive game that can be used to automatically assess behavioral attributes that are traditionally difficult to measure: like communication skills in difficult situations, group dynamics, and creative thinking.</p><p>Let’s streamline the behavioral interview process. Just as Hackerrank and CodeSignal transformed technical assessments, this could transform behavioral assessments.</p><p><strong>The Rules of the Game</strong></p><p>The game should be simple and easy to understand. So why not base the game off the board game ‘Codenames’. The rules are as follows: two teams compete by each having a ‘spymaster’ give one-word clues that can point to multiple words on the board. The other players on the team attempt to guess their team’s words while avoiding the words of the other team.” The team who uncovers the most words in the time slot wins.</p><p><strong>The Metrics</strong></p><p>So how can you assess behavior from a game of code names? What types of metrics could you create? What would be relevant sources of data? Consider the following:</p><ul><li>Chat Logs</li><li>User Demographics and Team Demographics</li><li>Transcripts of Video Chat Communication</li><li>Game Progress Throughout the Course of the Game</li><li>Changes in All the Above in response to more difficult game scenarios</li></ul><p>What type of metrics could one measure?</p><ul><li>Diversity Scores (Demographic data gathered for your team)</li><li>Tone and Attitude (Chat log and video chat communications text analysis)</li><li>Adaptation (Improvements in game outcomes over time)</li><li>Responses to Stress (Responses to curveball game scenarios)</li></ul><h3><strong>How is this Different from Existing Alternatives?</strong></h3><p><strong>Simplicity</strong></p><p>The game is easy to learn. Within minutes, one can get the hang of it.</p><p><strong>Variation</strong></p><p>Most games (and recruiting assessments) are easily game-able. This game isn’t. Taken from the English dictionary, there are billions of permutations of game outcomes.</p><p><strong>Multi-player, Live Dynamic Environments</strong></p><p>This game is dynamic, and changes based off each interaction. There’s no anticipating what comes next. You can only play off your team.</p><h3><strong>Why is it needed?</strong></h3><p>It does no good to hire a software engineer who is only adept in coding. A successful software engineer will need equally strong qualitative skills as technical skills. Behavioral interviews and reference checking are arguably the only way to gauge qualitative skills currently.</p><p>There are two problems I see with existing behavioral interviews:</p><p>1. Behavioral interviews ask mainly anecdotal questions that one can easily prepare for.</p><p>2. Behavioral interviews require significant manpower that firms often do not have.</p><h3><strong>Why a game?</strong></h3><p>Competitive, multiplayer games reveal characteristics that interviews cannot. It is no surprise that gaming companies like Riot Games run analytics from past chat logs and match histories as part of their behavioral screens.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=987954951f8a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[A Next-Generation Social Network For Volunteers]]></title>
            <link>https://xukevinwork.medium.com/a-next-generation-social-network-for-volunteers-75453f9d0b46?source=rss-711bee898a2c------2</link>
            <guid isPermaLink="false">https://medium.com/p/75453f9d0b46</guid>
            <category><![CDATA[volunteering]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[side-project]]></category>
            <category><![CDATA[social-media]]></category>
            <category><![CDATA[social-impact]]></category>
            <dc:creator><![CDATA[Kevin Xu]]></dc:creator>
            <pubDate>Fri, 14 Aug 2020 07:04:11 GMT</pubDate>
            <atom:updated>2022-02-03T17:30:30.969Z</atom:updated>
            <content:encoded><![CDATA[<h3>Volunteer Match 2.0</h3><h3>Inspiration</h3><p>I thought getting involved with volunteering would be easy.</p><p><em>I’m offering my time, skill, and effort for no cost. How hard can it be to find an organization that is a good fit?</em></p><p>Turns out it was super hard. I sent hundreds of emails, searched 10s of sites, and spent probably a few hours until I found an eventual gig.</p><p>Once I found the gig, it took weeks to get onboarded, and even once I was active, the registration and paperwork felt archaic.</p><p>I talked to my friends, organizations, and colleagues and I realized that I was not alone.</p><p><strong>Why is it so hard to get involved with volunteering?</strong></p><h3>Painpoints</h3><p>Inspired by my personal pain points as well as pain points from my direct network, here’s what I’ve noticed between volunteer-organization interactions.</p><ol><li><strong>It’s hard to find a “Match”</strong></li><li><strong>It’s even harder to find a Good Match</strong></li><li><strong>Communication with candidates takes a long time.</strong></li><li><strong>Onboarding for new candidates is slow</strong></li></ol><p>What if we made a platform that <strong>not only </strong>connected volunteers with organizations <strong>but also </strong>helped streamline their experience from <strong>discovery</strong> to <strong>activity </strong>to <strong>retention.</strong></p><p><strong>Discovery</strong></p><p>Organizations on our platform by default can set up</p><ul><li>Interest Forms</li><li>Info Sessions</li><li>Live Chats with representatives from the organizations</li><li>Metrics for tracking user engagement</li><li>Donation Widgets</li></ul><p>Before: send an email and wait for a response</p><p>Now: Engage on the platform, chat, and let the organization collect data on you.</p><p><strong>Onboarding:</strong></p><p>Organizations on the platform can</p><ul><li>Generate automated background check forms, W2s, payroll documents with DocuSign and other HRIS integrations</li><li>Generate ready-to-go itineraries with custom calendar exports for each new hire in one click of a button</li></ul><p><strong>Retention</strong></p><ul><li>Generate feedback forms for why a volunteer left in an instant</li><li>Find other similar organizations and share best practices and joint-sessions</li></ul><p>I imagine a world where this platform is a bigger and better volunteering match mixed with no-code and workflow tooling for volunteering organizations. Maybe, there is a possibility this could be simplified down to a flexible template for Volunteering organizations on a no-code website like Bubble.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=75453f9d0b46" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why Healthcare Management]]></title>
            <link>https://xukevinwork.medium.com/why-healthcare-management-f4452dc43133?source=rss-711bee898a2c------2</link>
            <guid isPermaLink="false">https://medium.com/p/f4452dc43133</guid>
            <dc:creator><![CDATA[Kevin Xu]]></dc:creator>
            <pubDate>Thu, 25 Jun 2020 05:25:38 GMT</pubDate>
            <atom:updated>2020-07-19T06:30:31.370Z</atom:updated>
            <content:encoded><![CDATA[<p>I’m a Computer Science and Healthcare Management major. When people hear me say this, they’ll often ask me why did you decide to study healthcare management? I don’t really have a personal story for it: no tale of a doctor saving a loved one like many of my classmates, no interest in potentially entering the field of medicine or management or consulting. I study healthcare management because of curiosity.</p><p>My thought process is more or less as follows: The United States spends upwards of three trillion dollars on healthcare a year. Healthcare is consistently one of the hottest topics in any political debate you’ll find on TV, and yet it’s something that I don’t understand at all. I’ll listen to politicians ramble for hours on hours, and sit there confused about what just happened. I can think of hundreds of questions that I have about our healthcare system. Why is the United States healthcare system so dang expensive? Why is the United States healthcare system so convoluted compared to other comparable countries? What’s a single payer system and what does it mean for how much I will have to pay? Why are healthcare outcomes so radically different across our states? Why do residents work such long hours, and how do they get feedback for their operations? These are all questions that I’ve asked at some points.</p><p>Not many universities in the United States offer a healthcare management department. Though I’m often the only engineer or frankly non pre-med in my classes, why not take these classes and try to get these questions answered?</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f4452dc43133" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>