<?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 Nathan Keolasy on Medium]]></title>
        <description><![CDATA[Stories by Nathan Keolasy on Medium]]></description>
        <link>https://medium.com/@keolazy1?source=rss-6d67c4cc4ca2------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*bTYlmo9opUf4M7jvKmh0WA@2x.jpeg</url>
            <title>Stories by Nathan Keolasy on Medium</title>
            <link>https://medium.com/@keolazy1?source=rss-6d67c4cc4ca2------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 14:34:52 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@keolazy1/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[Past Relationships w/ Brogramming — Cliches Included]]></title>
            <link>https://medium.com/@keolazy1/past-relationships-w-brogramming-cliches-included-2e4579dc6b11?source=rss-6d67c4cc4ca2------2</link>
            <guid isPermaLink="false">https://medium.com/p/2e4579dc6b11</guid>
            <category><![CDATA[advertising]]></category>
            <category><![CDATA[coding]]></category>
            <category><![CDATA[developer]]></category>
            <category><![CDATA[codingbootcamp]]></category>
            <category><![CDATA[marketing]]></category>
            <dc:creator><![CDATA[Nathan Keolasy]]></dc:creator>
            <pubDate>Sat, 22 Sep 2018 04:15:08 GMT</pubDate>
            <atom:updated>2021-09-04T00:28:28.820Z</atom:updated>
            <content:encoded><![CDATA[<h3>First Kiss</h3><p>Like most “90s-kids”, I grew up on the advent of social media platforms such as Myspace or Xanga (shoutout to you aZnBoii92’s out there). In these glory days, tons of people found themselves dabbling in HTML/CSS copy pasta, layouts, and song embeds. In a vain effort to better convey my teenage angst to internet strangers and middle-school crushes, I may have gotten more try-hard than most.</p><p>That was my first foray with “coding”. I used to think any more than that was reserved for basement-dwellers who wanted to make video games. Like dude, I just want to play Halo covered in Dorito crust, not make it.</p><h3>High School Prom Date</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/500/1*9pqi5QbJuLCYfqSQjAuNBg.jpeg" /></figure><p>In high school, my friend Alex had spoofed the district “Parent-Connection” site and tweaked some network settings at home. So in a way, he had it “automated” so his parents would arrive at a “dummy” site when checking his grades from home network. I know some of you sticklers may be wondering why someone with this genius would put so much effort into tricking his parents instead of just getting the good grades… but makes sense in hindsight. Cheers to all you beautiful minds out there.</p><p>That was the first time I realized how someone could use coding as a so-called “super-power” to help them with everyday life problems. I began to respect it more as I was able to fathom the possibilities that came with this power. So yeah Alex was a lvl 99 warlock and I was the edgy lvl 20 Chad Knight basically…</p><h3>Tinder Dates</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/1*3Wx-7tlvSTJQ9NbfwqFljA.jpeg" /></figure><p>Fast-forward to college days. My high-school friends are all split up around the country now. It’s also not socially acceptable to have all-nighters at a 16-XBOX LAN party anymore :(</p><p>We’re overly competitive, so we find this other avenue called “Daily Fantasy Sports”. In an effort to win more, I looked up what the “pros” did. Instead of relying on 7 chrome tabs to stay updated with all the latest stats &amp; player news, they would scrape desired data to an excel sheet and then make models out of them. I piggy-backed off some of their public code and made inferior models. Made some Coachella tickets. Not bad.</p><p>Pretty cool, but pretty no-contribution on my part. So yet another fling with coding, but zero-commitment.</p><h3>The one that got away…</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*04S09vFDbfgDpke7VwTE_w.jpeg" /></figure><p>So yeah there’s a point to all of these “cool stories”. After finishing my software engineering bootcamp with Galvanize, I found myself with a particular skillset. I could now program and deploy full stack applications using the latest frontend frameworks like React or Angular. If I had ZERO experience with a framework, I knew how to learn it quickly. Strangely enough, I still hadn’t the slightest clue of how to use it for “myself”.</p><p>I started working with my one of my old advertising startups again as I’m in the midst of that post-bootcamp job hunt stage (<strong><em>purgatory</em></strong>). After a week of catching up with new practices and general work, I started seeing things differently. Before, I had just accepted the daily creative process as is. Now, I was immediately noticing inefficiencies in some of our workflows. Some reoccurring processes seemed… well too reoccurring. I started writing potential fixes/solutions to make things more concise. Eventually, it felt less like work and more like a “rush”.</p><p>So now i’m writing a Medium post about it.</p><p>Yes, I’m that excited. (woah I can feel?)</p><p>Disclaimer: I hate that i’m always compelled to mention this, but by no means am I rambling about how excited I am cause “Bro I have sick app idea!”. I don’t care about product-to-market. Not point of story. Also this application is very specific to how my agency operates anyways.</p><h3>Bending the Knee: A lesson on ads you didn’t ask for.</h3><p>The purpose of this application is to better track A/B testing of creatives and copywriting across a media buying team. This prevents different media buyers on the same team from over-saturating the same “ad-combo” before it’s even proven to be effective. Providing a visual and quantifiable picture of what’s currently “Live” and being tested helps in decision-making for medium-large-sized teams and management. Lastly, this can inform the design team what variations they need to churn out next without having to remind them at 4 am in the morning (no outsourced language barriers needed yay!).</p><p>Collectively, this should all help in increasing the overall efficiency of advertising and operation costs.</p><p>A lot of factors go into whether an ad will be successful or not. You can theoretically have the greatest creative image ever with compelling copywriting, but if the algorithm or Life happens to catch you on a bad day (figuratively and quite literally — some products are more seasonal), you’ll be torching thru your advertising budget. That being said, it’s still better to stack the deck in our favor with as much <strong>methodology</strong> as possible.</p><p>Advertising agencies often have proven “winning formulas” they reuse and adjust slightly such as:</p><h4>Clickbait Intro → Model Showcase + Banner1 → Call to Action</h4><p>Due to the phenomenon known as “<strong>banner blindness</strong>”, even the most profitable 635% ROI ad will begin to have diminishing value after people have seen it around too much. Even unconsciously. The usual pivot to this is to keep the formula, but change the contents of the creatives.</p><h4>Same Cheesy Formula. Same Product. Different Content.</h4><p>In general, it’s good practice in variable testing to minimize the change of variables per run since it’ll be hard to deduce what attributed to either the success or failure of an “experiment” afterward.</p><p>Thanks, Baylor Biology.</p><p><strong>AdCross(AdX)</strong> aims to facilitate this process by generating different combinations of creatives and accurately tracking their usage. (<em>Yes. I came up with a cheesy name placeholder</em>)</p><h3>User Features:</h3><h4>Example:</h4><p>(3) different images for product, titled as such: [ Model1, Model2, Model3 ]</p><p>(1) banner for product, titled: [ Banner1 ]</p><p>(3) different CTAs for product, titled as such: [ cta1, cta2, cta3 ]</p><p><strong>AdCross(AdX)</strong> will begin by establishing this custom “<strong>formula</strong>” is composed of (3) parts.</p><p>It’ll then generate all the possible combos of these (9 combinations?) into different “<strong>Cards</strong>” similar to those on <strong><em>Trello</em></strong>.</p><h4>Generated Cards:</h4><p>Card1: [ ModelA — Banner1 — cta2 ] Card2: [ModelA — Banner1 — cta3]</p><ol><li>Users can move those cards to “<strong>Editing</strong>” for the video team to see and act on.</li></ol><p>2) Video team can drag to “<strong>Ready</strong>” column once videos are uploaded to company cloud.</p><p>3) Media Buyers can drag to “<strong>Live</strong>” column once Ad gets approved.</p><h4>Usage</h4><p>Each one of these columns will have a “<strong>Usage</strong>” section at the bottom. This will give a % stat of how much a certain entity is being used at each step of the process.</p><p><strong>Example:</strong></p><p>Editing Column: 100% ModelA, 100% banner1, 50% cta2, 50% cta3</p><p>Live Column: 50% ModelA, 100% banner1, 33% cta1 33% cta2, 33% cta3</p><p>This will be one of the base features I plan to roll out since it provides an actual statistic + visualization of what’s in play across different teams overall. Most of our inefficiencies mentioned before stem from this “<em>laissez faire”</em> approach of not accurately knowing if we’re running a certain “angle” into the ground.</p><h4>Assumptions:</h4><p>The application operates best under the assumptions that</p><ol><li>Agency has custom formulas that can be broken down to quantifiable parts (ie: 4-step)</li><li>Agency has strict naming conventions in play for file names.</li><li>Agency prefers drag-and-drop UI/UX opposed to boring excel sheet lol.</li></ol><h3>The Chase</h3><p>Hope you guys didn’t read too much of the last 20%. It’s especially dry and definitely more for me. I’ve been couped up in this work-at-home life for too long.</p><h4>If there’s one thing I hope you take away from this is… kudos to my relationship phase titles. They’re actually very d33p ok.</h4><p>Thanks for bearing with and getting this far in my brogrammer narrative. I’m gonna go like to Buckshots for the 834th time now because I hate myself.</p><p>Nathan Keolasy — (SEO PLEASE)</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2e4579dc6b11" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Q2-Backend Project Recap: Memoirs of Perpetual Imposter Syndrome.]]></title>
            <link>https://medium.com/@keolazy1/q2-backend-recap-524b4cab3f57?source=rss-6d67c4cc4ca2------2</link>
            <guid isPermaLink="false">https://medium.com/p/524b4cab3f57</guid>
            <category><![CDATA[web-development]]></category>
            <dc:creator><![CDATA[Nathan Keolasy]]></dc:creator>
            <pubDate>Thu, 19 Apr 2018 05:38:23 GMT</pubDate>
            <atom:updated>2018-04-20T16:05:58.676Z</atom:updated>
            <content:encoded><![CDATA[<p>Project Premise:</p><p>Goals &amp; Objectives:</p><ul><li>Present functional Full-Stack Web Application by deadline.</li><li>Exercise as much as AGILE development practices as a group as possible.</li><li>Understand how every component of the application works rather than only the features assigned to me.</li></ul><p><strong>4/13 — Friday</strong>: Instructors split up and assigned to teams to come up with our Quarter 2 project which requires both front-end and back-end (<strong>Full-Stack Development</strong>) skills we have learned so far.</p><p>Brainstormed several project ideas that would make use of the backend material we had been drilling in class. Naturally, this would be any application that would require usage of a database and <strong>Restful API</strong> .</p><p>Some initial ideas were “CRUD’s List”, “Small Business Employee Scheduler/Payroll”, and “Networking Match Finder”.</p><p><strong>4/14 — Saturday:</strong> Group came in early to get a jump-start on the project.</p><p>We drafted mockup pictures of all the “<strong>Views</strong>” (imagine every possible display view a user will see when navigating thru app, Home, Newsfeed, Notifications, etc) the app would have. We also mapped out <strong>Entity Relationship Diagrams (ERD)</strong> for how/where our data would exist in different tables &amp; how they would interact in <strong>MVP (Minimum Viable Product) </strong>form.</p><p>Got more experience in implementing <strong>SCRUM(Agile Development)</strong> project management and defining “user stories”. We spent the first session coding together, setting up boiler-plate code and the <strong>HTML/CSS</strong> template which would be used throughout mosts “Views”. We opted to go the “<strong>Git</strong> clone — authorize users” vs the “Git Fork — Open Source” style of <strong>version contro</strong>l work flow. This was the first time I was truly enabled to see the utility in these practices while working with a group. Merge conflicts galore.</p><p>At this time, I’m still not very confident at writing <strong>CRUD routes</strong> myself without constant reference to previous practice problems or assistance from instructors. In fear of being too dependent, I try to code the user-stories/features assigned to me using only online documentation as much as possible. Apparently hitting a brick wall is common practice around here.</p><p>8-hours and 34ish-Chrome tabs open later, brain fully fried and I find myself spacing out on why I’m looking at a certain tab in atom or chrome. Despite the exhaustion, I believe that I was able to transcend my previous limitation on understanding how <strong>Express</strong> Routing and <strong>Knex</strong> queries come together. Even if this new found confidence is a false reassurance, I have at the very least begun to memorize some general patterns in setup for whatever that’s worth.</p><p><strong>4/16 — Monday:</strong> Class resumes and the rest of the week is reserved for lab-time working on our respective group projects.</p><p>Thanks to our setup work on the weekend, I’m able to hit the ground running and systematically take on coding routes that different features of our app will require. As we slowly built our app out, we would discover new <strong>Features/User-stories</strong> needed that were not initially thought of.</p><p>By lunch I was pretty comfortable at writing routes for anything, so long as I took ample time racking my brain in visualizing how our database tables needed to interact with each other. Knowing how to use the appropriate <strong>table “joins”</strong> was initially challenging because I found myself referencing our ERD mockups and drawing <em>Venn diagrams</em> religiously (<em>often taking valuable laptop-screen real estate</em>). After getting past that learning curve, I requested to have more responsibility with coding the <strong>EJS views</strong> for the routes I wrote.</p><p>Initial Assumptions: I had this pre-disposition that I could always refine my HTML/CSS skills later and instead focus more on Javascript and other new concepts the program had to offer. Q2 came around and I quickly was at max bandwidth taking in all the new back-end concepts. Conclusion —</p><p><strong>Personal HTML/CSS Rating: 4.3/10</strong>.</p><p>I had some decent experience with using template literal variables in HTML from my front-end Q1 project. I assumed I’d be using .forEach() to populate a collection of data such as using it to render multiple row entries for features such as “My Events” or “People who RSVPd”.</p><p><strong>4/17-19 — Thursday: </strong>Geeze. Just when you think you’re at 90% MVP, new unsuspecting user stories surface to reality check you back to 75%.</p><p>We’ve finally hit MVP by the end of class. I’d really like to integrate the Eventbrite API. That’s currently at the top of my priority list in terms of stretch goals.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=524b4cab3f57" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>