<?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 Javier Segovia on Medium]]></title>
        <description><![CDATA[Stories by Javier Segovia on Medium]]></description>
        <link>https://medium.com/@_jotase?source=rss-64c23a1254a7------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*yUxSIinz9OI6uoB4GCKVig.jpeg</url>
            <title>Stories by Javier Segovia on Medium</title>
            <link>https://medium.com/@_jotase?source=rss-64c23a1254a7------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 17:48:23 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@_jotase/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[Coding is Old News — Mastering Management is the New Essential]]></title>
            <link>https://medium.com/@_jotase/coding-is-old-news-mastering-management-is-the-new-essential-a74724bab33f?source=rss-64c23a1254a7------2</link>
            <guid isPermaLink="false">https://medium.com/p/a74724bab33f</guid>
            <category><![CDATA[coding]]></category>
            <category><![CDATA[ideas]]></category>
            <category><![CDATA[entrepreneurship]]></category>
            <category><![CDATA[leadership]]></category>
            <category><![CDATA[management]]></category>
            <dc:creator><![CDATA[Javier Segovia]]></dc:creator>
            <pubDate>Mon, 03 Mar 2025 22:20:02 GMT</pubDate>
            <atom:updated>2025-03-05T11:57:05.811Z</atom:updated>
            <content:encoded><![CDATA[<h3>Coding? Nah, It’s All About Managing Now</h3><p>Hey there,<br> I’ve been asked so many times over coffee with friends, colleagues, and other professionals: “Should I learn to code? What programming language, framework, or tool should I pick up?” And honestly, my answer is pretty clear these days. With the rise of LLMs and the democratization of AI, the focus isn’t just on coding anymore. Instead, we need to learn how to <strong>manage</strong>.</p><p>I’ve spent years leading tech teams, developing AI, and automating just about everything. I’ve always been the person setting up scripts, bots, agents, or services to handle those repetitive tasks — both at work and in my personal projects. Back in the day, I relied heavily on my programming skills to get things done. But now? With agents becoming a reality and services popping up for even non-coders, everything’s a lot simpler. It’s only a matter of time before specialized agents emerge to tackle every repetitive task out there.</p><p>The real challenge isn’t the technology itself — it’s what happens when you have a big team, full of both people and digital agents, all waiting for direction. Without a leader who knows how to manage and empower them, you end up with an inefficient crew. Adding more talent without proper management just multiplies the chaos.</p><p>Imagine having an endless supply of digital assistants ready to handle the routine work. Without someone to effectively manage and coordinate them, you’re just creating another disorganized team. Sure, coding still has its place, but the real game-changer now is being able to turn your ideas into actionable plans and lead a team — be it human or digital — to execute them.</p><p>At the end of the day, it’s not about chasing the latest programming language or framework. It’s about mastering the art of management. That’s the key to scaling up, optimizing tasks, and truly making an impact in this new era of technology.</p><p>So next time someone tells you to learn to code, just say, “Nah, I’m busy managing the digital revolution.”</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a74724bab33f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Navigating Multilingual URLs: A Deep Dive into Best Practices]]></title>
            <link>https://medium.com/@_jotase/navigating-multilingual-urls-a-deep-dive-into-best-practices-33ef55e21cb8?source=rss-64c23a1254a7------2</link>
            <guid isPermaLink="false">https://medium.com/p/33ef55e21cb8</guid>
            <category><![CDATA[localization]]></category>
            <category><![CDATA[ux-design]]></category>
            <category><![CDATA[seo]]></category>
            <category><![CDATA[web-development]]></category>
            <category><![CDATA[programming]]></category>
            <dc:creator><![CDATA[Javier Segovia]]></dc:creator>
            <pubDate>Sun, 02 Mar 2025 16:40:27 GMT</pubDate>
            <atom:updated>2025-03-02T16:40:27.821Z</atom:updated>
            <content:encoded><![CDATA[<p>When you’re running a website for a global audience, one of the most underrated challenges is choosing the right URL structure for your different language versions. I recently dove into the nitty-gritty of the three major approaches — subdirectories, subdomains, and country-code top-level domains (ccTLDs) — and here’s what I learned. Spoiler: there’s no one-size-fits-all answer, but there are some clear pros and cons to consider.</p><h3>Subdirectories: One Domain, Many Languages</h3><p>Imagine all your content neatly housed under one roof. With subdirectories, you structure your URLs like this:</p><ul><li><strong>English:</strong> example.com/en/</li><li><strong>Spanish:</strong> example.com/es/</li></ul><p><strong>The Upside:</strong></p><ul><li><strong>Shared Authority:</strong> All language pages benefit from the link equity of your main domain. This means that your new localized content can gain traction faster because it’s riding on the coattails of your primary site’s SEO strength.</li><li><strong>Simpler Management:</strong> Everything is in one place. There’s no need to juggle multiple domains, DNS configurations, or separate CMS setups. This unified approach keeps things clean and cost-effective.</li><li><strong>Consistent Experience:</strong> Users get a cohesive brand experience. Switching languages is as simple as changing the folder in the URL, which reinforces a sense of continuity across your site.</li></ul><p><strong>The Downside:</strong></p><ul><li><strong>Subtle Geotargeting:</strong> The URL itself won’t scream “local” like a country-specific domain might. You’ll have to rely on hreflang tags and other signals to clarify which audience each section is for.</li><li><strong>Single Point of Failure:</strong> All languages share the same infrastructure. If something goes wrong on your main site, every language version could be affected.</li></ul><p>For most websites, subdirectories are the go-to option — they consolidate your SEO efforts and keep things streamlined. It’s a strategy that many large enterprises (think Apple or Samsung) have successfully adopted.</p><h3>Subdomains: Distinct Yet Connected</h3><p>Now, let’s consider subdomains. Here, you split your content across different subdomains:</p><ul><li><strong>English:</strong> en.example.com</li><li><strong>Spanish:</strong> es.example.com</li></ul><p><strong>The Upside:</strong></p><ul><li><strong>Flexibility and Separation:</strong> Subdomains can give each language version its own identity. This can be especially useful if you need to use different back-end systems, server locations, or even have different teams managing each language.</li><li><strong>Granular Control:</strong> With separate subdomains, you can more easily set up targeted SEO strategies or even geo-target each one independently in tools like Google Search Console.</li></ul><p><strong>The Downside:</strong></p><ul><li><strong>Split SEO Juice:</strong> One of the major drawbacks is that search engines may treat subdomains as separate sites. This means each subdomain has to earn its own authority from scratch — a process that can be slower compared to the consolidated strength of subdirectories.</li><li><strong>User Confusion:</strong> Not every visitor understands the nuance of subdomains. If someone lands on your main domain expecting to find localized content, they might not immediately realize they need to navigate to a different subdomain.</li></ul><p>Subdomains can work well if your localized content is drastically different or if you have the resources to build up each subdomain’s SEO independently. It’s a trade-off between flexibility and the challenge of splitting your link equity.</p><h3>ccTLDs: Going Local, Literally</h3><p>Then there’s the ccTLD approach — using country-code domains like:</p><ul><li><strong>United States:</strong> example.com</li><li><strong>Spain:</strong> example.es</li></ul><p><strong>The Upside:</strong></p><ul><li><strong>Powerful Local Signals:</strong> ccTLDs send a strong message to both users and search engines. A .es domain immediately signals that the content is tailored for Spain, which can boost local trust and rankings.</li><li><strong>Tailored User Experience:</strong> This method allows for complete localization. You can adjust content, pricing, and even design based on local market needs without any ambiguity.</li></ul><p><strong>The Downside:</strong></p><ul><li><strong>High Complexity and Cost:</strong> Managing multiple domains can get messy quickly. You’re looking at separate registrations, hosting setups, SSL certificates, and the ongoing challenge of maintaining brand consistency across each site.</li><li><strong>Diluted SEO Effort:</strong> Each domain is treated as a standalone entity. That means you have to build authority for each one from scratch — an approach that demands significant resources and time.</li></ul><p>ccTLDs make sense if you’re operating in markets where local presence is crucial and you have the budget to support multiple domains. They’re common among huge global players (think Amazon), but for most smaller businesses, the overhead can outweigh the benefits.</p><h3>SEO, Hreflang, and Beyond</h3><p>Regardless of your chosen URL structure, one thing remains non-negotiable: proper implementation of hreflang tags. These tags help search engines understand which version of your content to serve based on the user’s language and region. Whether you’re using subdirectories, subdomains, or ccTLDs, consistent hreflang usage is essential to prevent duplicate content issues and ensure the right pages appear in the right markets.</p><p>Also, consider your overall user experience. A clean, intuitive language selector and clear navigation between localized pages are as important as the URL structure itself. The goal is to provide a seamless experience where users feel like they’re exactly where they need to be — no confusing redirects or dead ends.</p><h3>So, What’s the Best Option?</h3><p>If you’re like me — seeking a balance between strong SEO performance and manageable maintenance — subdirectories are often the best starting point. They let you consolidate your SEO efforts and keep things simple, especially if you’re not running entirely separate businesses in different regions.</p><p>However, if your localized content requires distinct management or if you need to cater to very specific regional needs, then subdomains or even ccTLDs might be worth the extra effort. Just remember: with great flexibility comes greater responsibility. Each method demands its own set of SEO and technical considerations.</p><p>Ultimately, there’s no universal answer. Your choice should align with your business goals, available resources, and long-term strategy for international growth.</p><p>In the end, the key is to ensure that every visitor — no matter where they are — feels that your site speaks their language. And with the right URL structure, you can build a global presence that’s as robust in search rankings as it is in user experience. Happy optimizing!</p><p>Feel free to share your experiences or ask any questions in the comments below. Let’s keep the conversation going about building a truly international website.</p><p><em>If you enjoyed this deep dive, follow along for more insights on digital strategy and SEO best practices.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=33ef55e21cb8" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[AI Revolution: Embracing the Future and Using AI for Good]]></title>
            <link>https://medium.com/@_jotase/ai-revolution-embracing-the-future-and-using-ai-for-good-2ca1fa04c02a?source=rss-64c23a1254a7------2</link>
            <guid isPermaLink="false">https://medium.com/p/2ca1fa04c02a</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[productivity]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Javier Segovia]]></dc:creator>
            <pubDate>Fri, 06 Jan 2023 19:48:45 GMT</pubDate>
            <atom:updated>2023-01-06T19:49:57.877Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*-n9GYgRpQsGeknGg" /><figcaption>Photo by <a href="https://unsplash.com/@agk42?utm_source=medium&amp;utm_medium=referral">Alex Knight</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>Everybody is talking about ChatGPT but, the AI revolution is upon us and is changing the way we live and work in ways we could have never imagined. While it is natural to feel a little bit scared about the unknown, it is important to remember that AI has the potential to make our lives more efficient and easier in many ways.</p><p>Just like the industrial revolution and the development of electricity and the internet, the AI revolution will bring about both challenges and opportunities. It is up to us to decide how we want to use it.</p><h3>Automation</h3><p>One way that AI can be used to make our lives more efficient is through automation. Many repetitive and mundane tasks can be automated using AI, freeing up our time to focus on more meaningful and creative work. This can lead to increased productivity and job satisfaction.</p><h3>Data Analysis</h3><p>AI can also be used to analyze vast amounts of data and make predictions or recommendations that would be impossible for a human to do on their own. This can be especially useful in fields such as healthcare, finance, and marketing.</p><h3>Assistance</h3><p>Another way that AI can be used for the better is through its ability to assist with decision-making. By providing us with more information and different viewpoints, AI can help us make more informed and well-rounded decisions.</p><p>But it is important to remember that AI is not a replacement for human intelligence and judgment. It is simply a tool that can augment our abilities and help us make better decisions.</p><p>So how can we ensure that we are using AI ethically and for the benefit of society as a whole? One way is to have open and transparent discussions about the use of AI and its potential impacts. We should also consider the ethical implications of AI and ensure that it is developed and used in a fair and just way.</p><p>Another important aspect is to ensure that the development and use of AI are inclusive and benefit all members of society. This means considering the potential impacts on different groups of people and ensuring that the benefits of AI are shared fairly; For example, by democratizing education.</p><p>Yesterday I met an astronomer and asked him to explain to me “What existed before the big bang?” then, I used his explanation I asked to Chat GPT to explain the same thing if like I’m five in front of him, and his face was priceless, and it wasn&#39;t fear, it was the possibilities to spread the information to anyone without fuzzy words to the entire world that illuminated his face.</p><p>As with any technological advancement, there will be challenges and uncertainties. But by being open and proactive about the use of AI, we can ensure that it is a force for good in the world.</p><p>In conclusion, the AI revolution is upon us and it has the potential to greatly improve our lives and make us more efficient. While it is natural to feel a little bit scared about the unknown, it is important to remember that AI is simply a tool that can augment our abilities and help us make better decisions. By being open and proactive about the use of AI, we can ensure that it is used ethically and for the benefit of society as a whole.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2ca1fa04c02a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[My strategy with TradingAdvisor.app]]></title>
            <link>https://medium.com/@_jotase/my-strategy-with-tradingadvisor-app-6201e7ea1208?source=rss-64c23a1254a7------2</link>
            <guid isPermaLink="false">https://medium.com/p/6201e7ea1208</guid>
            <category><![CDATA[profit]]></category>
            <category><![CDATA[forex]]></category>
            <category><![CDATA[cryptocurrency]]></category>
            <category><![CDATA[investing]]></category>
            <category><![CDATA[trading]]></category>
            <dc:creator><![CDATA[Javier Segovia]]></dc:creator>
            <pubDate>Sat, 19 Feb 2022 19:11:46 GMT</pubDate>
            <atom:updated>2022-02-19T19:11:46.005Z</atom:updated>
            <content:encoded><![CDATA[<p>After years of experiences, losses, failures, books, and of course profits, I simplified everything I learned and made use of my expertise as a software engineer and data scientist at <a href="https://www.tradingadvisor.app/">https://www.tradingadvisor.app</a> If you want to know more about the reasons and how it works, I have an article where I talk about it <a href="https://medium.com/p/81e886b4f67e">https://medium.com/p/81e886b4f67e</a></p><p>That said, this is my day-to-day trading.</p><p><strong>1</strong>. When I have time (I have other projects), I go to <a href="https://www.tradingadvisor.app/">https://www.tradingadvisor.app/</a> and check which signals are live (Note: I only consider those with status BUY or SELL, those with HOLD is at your own risk if wants to enter late)</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*VjebzMdb1wXaYsO8WQyFzA.png" /></figure><p><strong>2.</strong> In this case, a currency has a buy signal (USD/SEK) so I proceed to check if it has any current or future news that may affect its volatility</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/988/1*OSv2_bqfLsU-fCO_t5Wjdg.png" /></figure><p>According to <a href="http://investing.com/">investing.com</a> there is nothing for the day, the news comes from the previous day, therefore I make the decision to trade this instrument.</p><p><strong>3. </strong>After deciding what to trade, the most important step is risk control, particularly I have this simple rule:</p><p><strong><em>The stop loss cannot represent more than 10% of my capital.</em></strong></p><p>That number will depend on each person, it is the one I use and it works for me because it is the risk I am willing to expose myself to. With the maximum stop loss established, now it remains to calculate the lot to operate.</p><p><strong>4.</strong> Operate and wait for results:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*C7AhixBICYZa0TTd7lIcAA.png" /></figure><p>Keep in mind what Take Profit 1 and Take Profit 2 mean:</p><ul><li>I use take profit 1 as the first level to reach, here I can make two decisions:</li></ul><ol><li>Establish a trailing stop loss: Recommended to get the maximum profit.</li><li>Close the operation: Recommended for safer profits.</li></ol><ul><li>I use take profit 2 as a take profit if only if, I have decided to use trailing stop loss on take profit 1, otherwise I don’t use it.</li></ul><p>As you can see, this is a trade on February 18, 2022 where I earned 400+ pips</p><p>Link to tradingview</p><p><a href="https://www.tradingview.com/chart/USDSEK/hYK7XgVk-USD-SEK-prediction-powered-by-tradingadvisor-app/">https://www.tradingview.com/chart/USDSEK/hYK7XgVk-USD-SEK-prediction-powered-by-tradingadvisor-app/</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6201e7ea1208" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to earn extra money in trading in a very easy and efficient way.]]></title>
            <link>https://medium.com/@_jotase/how-to-earn-extra-money-in-trading-in-a-very-easy-and-efficient-way-81e886b4f67e?source=rss-64c23a1254a7------2</link>
            <guid isPermaLink="false">https://medium.com/p/81e886b4f67e</guid>
            <category><![CDATA[fintech]]></category>
            <category><![CDATA[money]]></category>
            <category><![CDATA[trading]]></category>
            <category><![CDATA[hacking]]></category>
            <category><![CDATA[finance]]></category>
            <dc:creator><![CDATA[Javier Segovia]]></dc:creator>
            <pubDate>Tue, 11 Jan 2022 16:26:31 GMT</pubDate>
            <atom:updated>2022-01-11T16:26:31.848Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Uyk6A3RdIM_re9KW" /><figcaption>Photo by <a href="https://unsplash.com/@austindistel?utm_source=medium&amp;utm_medium=referral">Austin Distel</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h4>TL;DR All the strategies listed here are automated at <a href="https://www.tradingadvisor.app/">https://www.tradingadvisor.app</a> Free!!! I also recommend reading the full article to understand how it works before making the decision to use the tool.</h4><p>Do you want to make money by trading? Well, who doesn’t? It is a technology that has allowed us to perform a large number of activities that were previously available to a very small audience, and in this appears an activity that increasingly pursues more and more: making money by trading.</p><p>Knowing how to trade will open the door to financial freedom that you will never want to leave. Trading has become a very profitable tool for all those who want to manage their finances and are looking for options to invest their savings. Therefore, here you will find information about it and a simple (and almost automated) strategy that will help you make money at it.</p><h3>¿What is trade?</h3><p>According to <a href="https://www.investopedia.com/terms/t/trade.asp">Investopedia</a> “Trade is a basic economic concept involving the buying and selling of goods and services, with compensation paid by a buyer to a seller, or the exchange of goods or services between parties. Trade can take place within an economy between producers and consumers.”</p><p>The essence of trading is to make a good profit margin by predicting the movement in the price of any commodity or financial asset. For example, you can buy shares of a newly listed company at a very low price and then sell them when the price rises.</p><h3>What do I need to get started?</h3><p>If you are looking for some trading tips for beginners and know how to make your money more profitable, let me tell you that the rules work the same way regardless of your skill level, so before investing it is best to have enough knowledge to accept acceptable risks.</p><p>Although trading money sounds simple, it requires a bit of trading intuition. In addition, you will need to do some analysis to know when and where you are depositing your capital. Although it may sound like a gamble, not everything was left to chance.</p><p>In addition to skills and knowledge, you need an online broker to be able to perform your operations, I will not give recommendations to any because I do not advertise, and each one must be responsible for choosing the best according to their needs.</p><p>With the necessary skills, knowledge, and tools, then you must define strategies to start operating responsibly and not turn this into bets. In order to define strategies, you must then make an analysis, there are several types and here I will explain briefly what they are.</p><h3>Trading analysis</h3><p>There are two types of trading analysis you can perform: technical analysis and fundamental analysis.</p><h4>Technical analysis</h4><p>This type of trading analysis does not take into account factors related to fluctuations in the global economy but studies the patterns that may exist in the price movement of a particular financial asset.</p><p>Technical analysis could be categorized into:</p><p><strong>Quantitative analysis or indicators: </strong>In quantitative technical analysis, mathematical and statistical models are used to analyze the market and how prices move, including simple and complex calculations. The main inputs of this strategy are price and volume. Some indicators (the most commonly used) are:</p><ul><li>Moving Average (MA)</li><li>Exponential moving average (EMA)-</li><li>Stochastic oscillator.</li><li>Moving average convergence/divergence (MACD).</li><li>Bollinger Bands.</li><li>Relative Strength Index (RSI)</li><li>Fibonacci retracements.</li><li>Ichimoku Cloud.</li></ul><p><strong>Technical chart analysis:</strong> Chart analysis is a branch of technical analysis that only considers quotes. This is the purest technical analysis available and only pays attention to past price movements to create an approximate price of an asset in the future. Although in many cases chart analysis is confused with technical analysis, the truth is that technical analysis includes many other concepts.</p><p>When doing chart analysis, we are dedicated to finding patterns and figures that define the behavior of the market. Some of the patterns and figures to detect are:</p><ul><li>Rectangles</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/934/1*9IE7C-kZWPwAAG2iVfleBw.png" /></figure><ul><li>Flags</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uBXwUvnWwXYwIzXexg7k6Q.png" /></figure><ul><li>Head &amp; Shoulders</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9egsPXhBZAANG1iM2nlfSA.png" /></figure><p>And much more.</p><h3>Fundamental Analysis</h3><p>Fundamental analysis exceeds statistics and mathematics to focus on everything that affects prices, from social events to political speeches to ensure that the market value oscillates.</p><p>Knowing the macroeconomic data related to the production, export, and consumption of certain types of products can have a significant impact on the price of an asset at any given time, so you can know when to invest.</p><p>Furthermore, by analyzing these supply and demand factors, traders based on the basic approach can identify profitable trades that most technical strategies would miss and merit, and this generates profits by trading efficiently.</p><p>In the end, the information provided by fundamental analysis is usually better in terms of risk management. If the trader knows why the price is going up or down, he can analyze the trade from an emotional perspective and determine whether it will benefit his portfolio.</p><h3>Ok, but let’s talk about strategies, how can I perform well in trading?</h3><p>Having mentioned all of the above, it should be understood that there are different types of financial instruments, each with its nature, market hours, behavior, etc. These are:</p><ul><li>Commodities</li><li>Currencies</li><li>Cryptocurrencies</li><li>Indices</li><li>Stocks</li><li>Bonds</li><li>ETFs</li><li>Funds</li><li>Certificates</li></ul><p>I have several strategies covering commodities, currencies, cryptocurrencies, indices, and stocks. All strategies are similar in form, they only differ in indicators and patterns that I highlight for each type of instrument, and the strategy is as follows:</p><ol><li>I have a list of instruments to follow, <strong>38 currencies</strong> (the major ones) <strong>3 cryptocurrency pairs</strong> (btc/usd, btc/eur, eth/usd) <strong>30 indices, 34 commodities, and more than 10000 stocks!</strong></li><li>For each of the instruments (more than 10k), I analyze their trends (bullish and bearish) and select those where their trends are very noticeable based on their past performance in different time intervals (Quantitative Analysis).</li><li>From the selected instruments, I apply the calculations of several indicators (some special for each type of instrument) and filter by those that validate the trend of the same (Indicator Analysis).</li><li>With the result of the previous filters, then I start looking for patterns and figures that also validate the trends of the instruments already filtered (Graphical analysis).</li><li>With a reduced list of instruments with validated trends, I calculate the supports and resistances and based on that, I calculate the pivot points of each one using several intervals, and with the different techniques (Floor, Camarillo, Fibonacci, Demarks, and Woodie) and I calculate the best targets to define my stop loss and take profit.</li><li>Then I check that the relation between losing and winning is favorable, that is to say, I do not operate in an instrument where I can win 1 if I can lose 10, I try that the minimum is 3/7, and those that do not I eliminate them from the list.</li><li>So far, out of the more than 10000 instruments, I would have less than 20 instruments to look at (even less, I am always improving more and more the filtering process) and then for each one I review relevant news or future announcements that may affect the behavior of each one of them (Fundamental Analysis), to then make a last manual filter and stay with the most “safe” ones and after that, decide to trade.</li></ol><h4>How to review more than 10k instruments and do all these steps is an easy way?</h4><p>Well, using my programming and data science skills I do all this analysis in a matter of seconds so that the calculated signal does not expire at the moment of finishing the analysis. That is why I have created <a href="https://www.tradingadvisor.app/">https://www.tradingadvisor.app</a> and saved myself all this work that I have automated with years of experience and learning. Professional traders (I am not, I am a full-time software engineer, and this is one of my hobbies/source of additional income) are dedicated only to a handful of instruments which they can learn their behavior and are aware 24/7 doing analysis and reviewing all global events that may affect their investments. In my case, on my breaks I check what signals I have detected from the analysis of thousands of instruments, and I make sure that there is no news on the day that can damage the forecast of the same, and <em>voila</em>!</p><h4>¿How can I start using <a href="https://www.tradingadvisor.app/">https://www.tradingadvisor.app</a> ?</h4><p>Simple, I recommend you first use a demo account (fictitious money) of the broker of your choice and do the following:</p><ol><li>Check that the last signal is not too old, they are usually renewed every 15 to 30 minutes, but if not, this is an indicator that the signals displayed are not as reliable.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uxS6Lttvb8NKuoh3oyPo2Q.png" /></figure><p>2. Check if any of the instruments with which you want to trade, do not have any close news that may affect its performance (I have this in my TO DO list; add a visual indicator). There are several applications with this information, for example, <a href="https://tradingeconomics.com/calendar">https://tradingeconomics.com/calendar</a>.</p><p>3. Once you have selected the instruments with their corresponding stop loss and take profit, and you have validated that there will be no news that can change their trend, then you only need to buy or sell at your trusted broker :).</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=81e886b4f67e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Como gano dinero extra en trading de manera muy fácil y eficiente.]]></title>
            <link>https://medium.com/@_jotase/como-gano-dinero-en-trading-de-manera-muy-f%C3%A1cil-y-eficiente-3ae8204d0503?source=rss-64c23a1254a7------2</link>
            <guid isPermaLink="false">https://medium.com/p/3ae8204d0503</guid>
            <category><![CDATA[money]]></category>
            <category><![CDATA[trading]]></category>
            <category><![CDATA[apps]]></category>
            <category><![CDATA[hacking]]></category>
            <category><![CDATA[finance]]></category>
            <dc:creator><![CDATA[Javier Segovia]]></dc:creator>
            <pubDate>Mon, 10 Jan 2022 01:52:42 GMT</pubDate>
            <atom:updated>2022-01-11T13:23:31.779Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Uyk6A3RdIM_re9KW" /><figcaption>Photo by <a href="https://unsplash.com/@austindistel?utm_source=medium&amp;utm_medium=referral">Austin Distel</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h4>TL;DR Todas las estrategias acá plasmadas están automatizadas en <a href="https://www.tradingadvisor.app/"><strong>https://www.tradingadvisor.app</strong></a> Gratis!! De igual forma recomiendo leer el articulo completo para comprender como funciona antes de tomar la decisión de usar la herramienta.</h4><p>¿Quieres ganar dinero con el Trading? Bueno, ¿quién no? Es la tecnología la que nos ha permitido realizar una gran cantidad de actividades que antes estaban al alcance de un público muy reducido, y en esta aparece una actividad que cada vez persigue más: ganar dinero por el método transaccional.</p><p>Saber operar te abrirá la puerta a la libertad financiera de la que nunca querrás salir. De hecho, el trading se ha convertido en una herramienta muy rentable para todos aquellos que quieren administrar sus finanzas personales y buscan opciones para invertir sus ahorros. Por lo tanto, acá encontraras información al respecto y una estrategia simple (y casi automatizada) que te ayudara a ganar dinero en esto</p><h3>¿Qué es el ‘trading’?</h3><p>Según <a href="https://www.investopedia.com/terms/t/trade.asp">Investopedia</a> “El ‘trading’ es un concepto económico básico que implica la compra y venta de bienes y servicios, con una compensación pagada por un comprador a un vendedor, o el intercambio de bienes o servicios entre las partes.”</p><p>La esencia del trading es obtener un buen margen de beneficio al predecir el movimiento en el precio de cualquier producto o activo financiero. Por ejemplo, puede comprar acciones de una empresa nueva que cotiza en bolsa a un precio muy bajo y luego venderlas cuando suba el precio.</p><h3>¿Qué se necesita para empezar?</h3><p>Si está buscando algunos consejos comerciales para principiantes y sabe cómo hacer que su dinero sea más rentable, déjeme decirle que las reglas funcionan de la misma manera independientemente de su nivel de habilidad, por lo que antes de invertir es mejor tener suficiente conocimiento para aceptar riesgos aceptables.</p><p>Aunque operar con dinero suena simple, necesita un poco de intuición comercial. Además, necesitará hacer un cierto análisis para saber cuándo y dónde está depositando su capital. Aunque pueda parecer una apuesta, no todo se dejó al azar.</p><p>Además de habilidades y conocimientos, necesitas un broker online para poder realizar tus operaciones, no daré recomendación de ninguno porque no hago publicidad, y cada quién debe ser responsable de elegir el mejor según sus necesidades.</p><p>Con las habilidades, conocimientos y herramientas necesarias, entonces se deben definir estrategias para comenzar a operar de manera responsable y no convertir esto en apuestas. Para definir estrategias se debe entonces hacer análisis, existes diversos tipos y acá les explico brevemente de que se trata.</p><h3>Análisis de trading</h3><p>Existen dos tipos de análisis de trading que puedes realizar: el análisis técnico y el análisis fundamental.</p><h4>Análisis Técnico</h4><p>Este tipo de análisis de trading no toma en cuenta factores relacionados con las fluctuaciones en la economía global, pero estudia los patrones que pueden existir en el movimiento de precios de un activo financiero en particular.</p><p>Los análisis técnicos podrían categorizarse en:</p><p><strong>Análisis cuantitativo o indicadores: </strong>En el análisis técnico cuantitativo, se utilizan modelos matemáticos y estadísticos para analizar el mercado y cómo se mueven los precios, incluidos cálculos simples y complejos. Los principales insumos de esta estrategia son el precio y el volumen. Algunos indicadores (los mas utilizados) son:</p><ul><li>Media móvil (MA)</li><li>Media móvil exponencial (EMA)</li><li>Oscilador estocástico.</li><li>Convergencia/divergencia de medias móviles (MACD)</li><li>Bandas de Bollinger.</li><li>Índice de fuerza relativo (RSI)</li><li>Retrocesos de Fibonacci.</li><li>Nube de Ichimoku.</li></ul><p><strong>Análisis técnico de gráficos: </strong>El análisis de gráficos es una rama del análisis técnico que solo considera cotizaciones. Este es el análisis técnico más puro disponible y solo presta atención a los movimientos de precios pasados ​​para crear un precio aproximado de un activo en el futuro. Aunque en muchos casos el análisis de gráficos se confunde con el análisis técnico, lo cierto es que el análisis técnico incluye muchos otros conceptos.</p><p>Al hacer análisis de gráficos, nos dedicamos a encontrar patrones y figuras que definen el comportamiento del mercado. Algunos de los patrones y figuras a detectar son:</p><ul><li>Rectángulos</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/934/1*9IE7C-kZWPwAAG2iVfleBw.png" /></figure><ul><li>Banderas</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uBXwUvnWwXYwIzXexg7k6Q.png" /></figure><ul><li>Hombro Cabeza Hombro</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9egsPXhBZAANG1iM2nlfSA.png" /></figure><p>Y muchos otros más.</p><h3>Análisis Fundamental</h3><p>El análisis fundamental excede las estadísticas y las matemáticas para centrarse en todo lo que afecta a los precios, los eventos sociales para discursos políticos garantizan que el valor de mercado oscila.</p><p>Conocer los datos macroeconómicos relacionados con la producción, exportación y consumo de ciertos tipos de productos puede tener un impacto significativo en el precio de un activo en un momento determinado, por lo que puede saber cuándo invertir.</p><p>Además, al analizar estos factores de oferta y demanda, los operadores basados ​​en el enfoque básico pueden identificar operaciones rentables que la mayoría de las estrategias técnicas perderían y merecerían, y esto genera ganancias al operar de manera eficiente. .</p><p>Al final, la información proporcionada por el análisis fundamental suele ser mejor en términos de gestión de riesgos. Si el operador sabe por qué el precio sube o baja, puede analizar la operación desde una perspectiva emocional y determinar si beneficiará a su cartera.</p><h3>Ok, pero hablemos de estrategias, ¿Cómo puedo tener buen rendimiento al hacer trading?</h3><p>Mencionado todo lo anterior, se debe entender que hay distintos tipos de instrumentos financieros cada cual con su naturaleza propia, horarios de mercados, comportamientos, etc. Estos son:</p><ul><li>Materias primas</li><li>Divisas</li><li>Criptomonedas</li><li>Indices</li><li>Acciones</li><li>Bonos</li><li>ETFs</li><li>Fondos</li><li>Certificados</li></ul><p>Tengo varias estrategias que abarcan las materias primas, divisas, criptomonedas, indices y acciones. Todas las estrategias son similares en forma, solo difieren en indicadores y patrones que resalto para cada tipo de instrumento, y la estrategia es la siguiente:</p><ol><li>Tengo una lista de instrumentos a seguir, <strong>38 divisas</strong> (las principales) <strong>3 pares de criptomonedas</strong> (btc/usd, btc/eur, eth/usd) <strong>30 indices</strong>, <strong>34 materias primas, y más de 10000 acciones!</strong></li><li>Para cada uno de los instrumentos (mas de 10000) analizo sus tendencias (alcistas y bajistas) y filtro aquellos donde sus tendencias son muy notorias en base a sus rendimientos anteriores en distintos intervalos de tiempo (Análisis cuantitativo) .</li><li>De los instrumentos filtrados, le aplico los cálculos de varios indicadores (algunos especiales para cada tipo de instrumento) y filtro por aquellos que validan la tendencia de los mismos (Análisis de indicadores).</li><li>Con el resultado de los filtros anteriores, entonces comienzo a buscar los patrones y figuras que también validen las tendencias de los instrumentos ya filtrados (Análisis gráfico).</li><li>Ya con una lista reducida de instrumentos con tendencias validadas, calculo los soportes y resistencias y en base a eso, calculo los pivot points de cada uno utilizando varios intervalos y con las distintas técnicas (Floor, Camarillo, Fibonnaci, Demarks, y Woodie)y calculo los mejores target para definir mi stop loss y mis take profit</li><li>Luego reviso que la relación entre perder y ganar sea favorable, es decir, no opero en un instrumento donde puedo ganar 1 si puedo perder 10, trato de que lo mínimo sea 3/7 , y los que no los elimino de la lista.</li><li>Hasta el momento de los mas de 10000 instrumento, me quedarían en la lista menos de 20 instrumento que mirar (inclusive menos, siempre voy blindando cada vez mas el proceso de filtrado) y entonces por cada uno reviso noticias relevantes o anuncios futuros que puedan afectar el comportamiento de cada uno de ellos (Análisis fundamental), para así entonces hacer un ultimo filtro manual y quedarme con los mas “Seguros” y luego de eso, decidir operar.</li></ol><h4>¿Cómo revisar más de 10000 instrumentos y hacer todos estos pasos es una manera fácil?</h4><p>Pues usando mis habilidades de programación y ciencia de datos hago todo este análisis es cuestiones de segundos, de manera que, la señal calculada no expire al momento de terminar el análisis. Por ello he creado <a href="https://www.tradingadvisor.app/">https://www.tradingadvisor.app</a> y me ahorro todo este trabajo que llevo automatizado con años de experiencia y aprendizaje. Los traders profesionales (yo no lo soy, soy ingeniero de software a tiempo completo, y este es uno de mis hobbies/fuente de ingreso adicional) Se dedican solo a un puñado de instrumentos la cual puedan aprender su comportamiento y están al tanto 24/7 haciendo análisis y revisando todos los acontecimientos globales que puedan afectar tus inversiones. Yo en otro caso, en mis breaks reviso que señales he detectado de análisis de miles de instrumentos, y me percato de que no haya ninguna noticia en el día que pueda dañar el pronostico de los mismos y voila!</p><h4>¿Cómo puedo comenzar a usar <a href="https://www.tradingadvisor.app/">https://www.tradingadvisor.app</a> ?</h4><p>Sencillo, te recomiendo primero usar una cuenta de prueba (dinero ficticio) del broker que escojas y hacer lo siguiente:</p><ol><li>Revisar que la ultima señal no tenga mucho tiempo de antigüedad, por lo general se renuevan cada 15 o 30 minutos, pero de ser el caso de que no, esto es un indicador de que las señales mostradas no sean de tanta fiabilidad</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uxS6Lttvb8NKuoh3oyPo2Q.png" /></figure><p>2. Revisar si alguno de los instrumentos con los cuales quieras operar, no tenga alguna noticia cercana que pueda afectar su rendimiento (tengo en mí TO DO list, agregar un indicador visual para esto). Existen varias aplicaciones con esta información, ejemplo <a href="https://tradingeconomics.com/calendar">https://tradingeconomics.com/calendar</a></p><p>3. Ya con instrumentos seleccionados con su stop loss y take profit correspondiente, y además, haber validado de que no habrá ninguna noticia que pueda cambiar su tendencia, entonces solo necesitas comprar o vender en tu broker de confianza :).</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3ae8204d0503" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Easy hack for procrastinators that really works! (at least for me)]]></title>
            <link>https://medium.com/@_jotase/easy-hack-for-procrastinators-that-really-works-at-least-for-me-fb2777e98037?source=rss-64c23a1254a7------2</link>
            <guid isPermaLink="false">https://medium.com/p/fb2777e98037</guid>
            <category><![CDATA[self-awareness]]></category>
            <category><![CDATA[productivity]]></category>
            <category><![CDATA[personal-development]]></category>
            <category><![CDATA[self-improvement]]></category>
            <category><![CDATA[life-hacking]]></category>
            <dc:creator><![CDATA[Javier Segovia]]></dc:creator>
            <pubDate>Sun, 09 May 2021 03:36:20 GMT</pubDate>
            <atom:updated>2021-05-09T03:36:20.525Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*P9o59nAmsKWvW0UR" /><figcaption>Photo by <a href="https://unsplash.com/@brett_jordan?utm_source=medium&amp;utm_medium=referral">Brett Jordan</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>Have you been in this situation where you have a lot of things to do? Still, you prefer to do something else non-productive. You are so aware that you’re so wrong about your acting but, there is a divine force that makes you procrastinate and says, “It’s all right, you can do that later. It’s better to gain some strengths and energy to accomplish that task anyway, and you don’t have it right now, do you? So let’s relax a little bit watching for the N-th time this episode of <em>The Office,</em> and you will back ready for action!”. Guess what? That’s B****ht.</p><h3>First. Let&#39;s understand the problem</h3><h4>What is procrastination?</h4><p>Procrastination is the act of delay or postponing your duties because a divine force won’t let you behave appropriately to your responsibilities. This behavior is so typical nowadays. In fact, has been common for centuries. The ancient greeks philosopher has a word this, and is <em>Akrasia</em> “is described as a lack of self-control or the state of acting against one’s better judgment.”</p><h4><strong>Why do we procrastinate?</strong></h4><p>It could be two things; you’re anxious and a bunch of incomplete tasks overwhelming you, or you’re an instant gratification junkie.</p><p>Having a lot of pending tasks is an overwhelming situation, and worse if you’re an anxious person, and that’s because you will need to complete all of them simultaneously. So it’s probable that you start to overthink the solutions, prioritize what can be or cannot be done, the costs, the danger of not completing a particular task successfully, and so on. And that, my friends, it’s paralyzing, frustrating, and exhausting! So before you start anything, you burned all your energy already.</p><p>If you’re an instant gratification junkie, you will prefer something like watching that funny episode of <em>The Office </em>that will make you laugh hard in the first seconds of the opening (Instantly), instead of doing the daily workout that will prevent a stroke to the future you, your older self (considering you get old and will not be killed by a bus tomorrow, I hope not). And that makes sense, right? So, why the hell would I prefer to do something that will benefit my future self in I-don’t-know-when-and-I-don’t-want-to-wait time instead of doing something pleasurable that will provide satisfaction instantly to me? To the self that lives in the present. This reminds me of a quote from one of the greatest of our age, Homer Simpsons <strong>“That’s a problem for future Homer. Man, I don’t envy that guy.</strong>”</p><h4>Are you saying there are two me&#39;s?</h4><p>Yes!, well, there are more than oneself; if we think in time perspective, there are infinite future self and a finite number of past selves, but to keep it simple, just think that we are a past, present, and future self, and we have to take care of all three. Besides those guys, there is a little one living inside our head, and it is the …(<em>mystery</em> sound effect) Divine Force!</p><h4>Who is this divine force?</h4><p>To be honest, there is no such thing; that force that makes sexy proposals to us is the better salesman that we can encounter, and it’s us! Let’s think about that for a minute. What are the best qualities a salesman should have? I can think of three of them:</p><ul><li><strong>Ability to listen:</strong> To succeed in a sale, you have to know what your client want/need, in that case, you will know what to offer that makes you earn a profit.</li><li><strong>Empathy: </strong>Knowing how your client feels, putting in his shoes, gives you massive leverage in a negotiation. In fact, that leverage is that huge that you can simply sell the most expensive option if you know how to use it.</li><li><strong>Ability to improve his strategies: </strong>The sales world is constantly changing, so a salesman must prepare and improve himself to fit whatever situations he had and make that sale!</li></ul><p>We’re talking about this salesman with that qualities and more because; he doesn’t need to listen to you because you can read your mind. He doesn’t need empathy because he knows exactly how you feel. He constantly changes with you, so his strategies will always work!</p><h3>So, how can I fix this thing?</h3><h4>Don’t think, just do</h4><p>As I mentioned before, thinking of all the tasks you need to get done is exhausting! So, try to avoid prioritizing, calculate costs or whatever action that not necessarily it’s getting you to complete your duties. Just pick one randomly if you want and do it. Desmond Tutu once wisely said that “there is only one way to eat an elephant: a bite at a time.”</p><h4>KISS</h4><p>Keep it simple stupid. This principle is a design rule that states that systems perform best when they have simple designs rather than complex ones. So, if one of your tasks is big or complicated, divide it into smaller and simple tasks.</p><h4>Be gentle to yourself.</h4><p>If you’re struggling to succeed in something, don’t punish yourself. It is already hard to avoid procrastinating, so you will make it worse if you feel bad. And be honest; if the thing you’re doing is impossible, then just let it go, and pick another one of your tasks, but if it is mandatory, remember the KISS principle.</p><h4>Don’t forget to congratulate and reward yourself.</h4><p>This is the most important thing you have to do, especially for instant gratification junkie. After accomplishing one of your tasks, congratulate and reward yourself. This is a shot of dopamine; the same you got when you scroll your social media and found out a cool meme, but 10 times better. This creates and reinforces your habit of getting things done.</p><h4>Compare to your past self.</h4><p>Every time you notice that you are becoming more productive than before, try to remember how awful you were being a procrastinator, and this is dope. You will feel powerful being aware that you will look forward to knowing how it will be your future self at this pace.</p><h3>Wrapping up: A practical solution</h3><p>Now you here. I will share with you the little hack I developed to myself that works like a charm. It’s straightforward; just write down in a post-it pack (don’t remove from the rest) paper by paper anything you think at the moment that has to be done. Limit your number of tasks to think about. Use FIFO (First in, first out) criteria to solve them. After solving any task, go open a bottle of champagne and keep visible your victories. Easy right?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/1*DKvG1VmtZWvRunz_NmA0Sg.jpeg" /></figure><p>Now let explain what does work for me, step by step.</p><ul><li><strong>Write down in a post-it pack (don’t remove from the rest) the tasks paper by paper:</strong> It helps me avoid unnecessary stress, just knowing that all the things I have to do are written on a piece of paper. They won’t be forgotten.</li><li><strong>Limit your number of tasks to think about: </strong>I allow myself to think of a maximum of 8 things to solve simultaneously. This is a trick because it’s the magic number for me to keep me aware without getting overwhelmed. I keep my monitor always with 8 post-its on it that are impossible to ignore because I see them the most part of my day. This made me think only about those 8 tasks, and I really forgot about the 8273492387 more that I have in queue.</li><li><strong>Use FIFO (First in, first out) criteria to solve them: </strong>This simple way of selecting my tasks to get done prevents me from overthinking and trying to prioritize. I just pick the first in the queue of post-its and move it to my monitor when a spot is available.</li><li><strong>After get things do, celebrate!: </strong>This is the most important action you have to do, even more than solve your task. If you celebrate every time you complete one of your duties, you will feel so great (hi dopamine!), and it’s addictive! You will be thinking about the next celebration. If you have the time, you’ll want to take another post-it immediately. The way you celebrate and reward yourself is up to you. I see myself like Michael Jordan flying and making a dunk, grabbing my piece of paper and putting it in a jar.</li><li><strong>Keep visible your victories: </strong>I used a glass to put the piece of paper of the completed task. Every day I see two things: My 8 piece of paper waiting to get picked on my monitor and the glass getting full of post-its of completed tasks. This motivates me a lot! All I want to do is fill that glass of post-its, it doesn’t make sense, but god, it’s really gratifying. Btw I have to change my glass by a jar sometime after.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/720/1*VB9W2G1Z7oGos_l-0IjruQ.jpeg" /></figure><p>It’s essential to know what makes you procrastinate; I think it’s the most important thing. If you’re capable of that, then you can figure out how to stop procrastination. That’s precisely what I did, and it helps me to try different things to fix that.</p><p>I hope this little trick be the solution for you and saves you time experimenting, but I recommend you that experiment anyway. Let me know what you do to stop procrastination and if this “hack” helps you, and if it does, and you modify it to make it better, please, share it!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fb2777e98037" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Afilar el hacha]]></title>
            <link>https://medium.com/@_jotase/afilar-el-hacha-f5fe5be7dfb6?source=rss-64c23a1254a7------2</link>
            <guid isPermaLink="false">https://medium.com/p/f5fe5be7dfb6</guid>
            <category><![CDATA[crecimiento-personal]]></category>
            <category><![CDATA[lifehacks]]></category>
            <category><![CDATA[liderazgo]]></category>
            <category><![CDATA[leadership]]></category>
            <category><![CDATA[software-engineering]]></category>
            <dc:creator><![CDATA[Javier Segovia]]></dc:creator>
            <pubDate>Tue, 21 Jul 2020 19:13:57 GMT</pubDate>
            <atom:updated>2020-07-21T22:18:51.771Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*fungjn8qI5ka3l_0" /><figcaption>Photo by <a href="https://unsplash.com/@jabdilla_creative?utm_source=medium&amp;utm_medium=referral">Jason Abdilla</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p><em>En cierta ocasión un joven llegó a un campo de leñadores con el propósito de obtener trabajo, con un gran entusiasmo decide hablar con el responsable del lugar, y este al ver su aspecto, motivación y fortaleza lo contrata sin pensarlo y le pide comenzar al siguiente día.</em></p><p><em>Durante su primer día en la montaña, el joven trabajó duramente y llegó a cortar muchos árboles.</em></p><p><em>Al segundo día, trabajó tan duro como el primero pero, su producción fue escasamente la mitad del primer día.<br> <br> El tercer día, el joven frustrado por su desempeño del día anterior, decide trabajar el doble de duro con el objetivo de conseguir una producción mucho mejor que la del primer día, pero este solo alcanza una producción casi nula.</em></p><p><em>Cuando el leñador jefe se da cuenta del bajo rendimiento del joven, se le acerca y le pregunta “¿Cuándo fue la última vez que afilaste tu hacha?”. El joven un poco molesto y confundido le responde; “No tengo tiempo para ello, he estado muy ocupado cortando árboles”.</em></p><p>Este es el último y más importante hábito mencionado por <a href="https://g.co/kgs/eFnG7w">Stephen Covey</a> en su libro <a href="https://es.wikipedia.org/wiki/Los_siete_h%C3%A1bitos_de_la_gente_altamente_efectiva">“Los 7 hábitos de la gente altamente efectiva”</a> donde explica la importancia de preparar y mantener el activo más importante que todos tenemos; Nosotros mismos. Es el hábito más importante porque se basa la capacidad que tenemos para renovarnos y fortalecernos tanto física, mental y espiritualmente. Sin embargo este hábito no se limita a nosotros solamente, ya que, nosotros somos la base de nuestro propio universo y sobre ello, coexisten varias dimensiones, escenarios y vidas que se ven afectadas por la fuerza y estabilidad de las mismas, así como una edificación depende de los cimientos del mismo para mantenerse.</p><p>He querido en este artículo listar y detallar algunos de los ámbitos más allá del personal donde este hábito es altamente efectivo, los cuales son:</p><ul><li>Personal</li><li>Interpersonal</li><li>Profesional</li></ul><h3>Ámbito Personal</h3><p>Tal como menciona <a href="https://g.co/kgs/eFnG7w">Stephen Covey</a> existen tres dimensiones en la cual debemos ocuparnos para mantener el hacha (YO) bien afilada, y estas son:</p><h4><strong>Dimensión Física</strong></h4><p>Hace referencia a nuestro cuerpo y a nuestra salud, acá debemos reflexionar acerca de nuestro:</p><ul><li><strong>Estado de salud:</strong> <br>- ¿Estamos consciente de nuestro estado de salud?<br>- ¿Con qué regularidad me hago exámenes de salud y/o voy al odontólogo?<br>- Si padezco una enfermedad ¿Estoy haciendo lo correcto y/o necesario para erradicarla o controlarla?</li></ul><p>Padecer alguna enfermedad es un estado que a muchas personas le pueden limitar en sus quehaceres diarios, así como en sus proyectos futuros, por eso, es sumamente importante hacerse responsable de esto para evitar futuras preocupaciones.</p><ul><li><strong>Estado físico: <br></strong>-<strong> </strong>¿Nos ejercitamos lo suficiente?<br>- ¿Estamos físicamente en la forma que queremos?<br>- ¿Nuestra higiene es la adecuada?<br>- ¿Nos preocupamos por cómo nos vemos?<br>- ¿Con qué frecuencia practicamos algún deporte?<br>- ¿Duermo lo suficiente?</li></ul><p>Hacer ejercicios y/o practicar algún reporte de manera constante, es muy importante a cualquier edad; ayuda a prevenir enfermedades, mejorar el humor, aumentar el autoestima, y junto con el autocuidado se logra disminuir la posibilidad de desarrollar trastornos como la depresión y la ansiedad. No es necesario dedicar horas al día para esto, conmutando a un establecimiento, haciendo filas o esperar para poder entrenar, existen varias alternativas como ejercitarse en casa 15 minutos, asegurarse de caminar unos 30 en el transcurso del día, etc.</p><ul><li><strong>Ingesta:<br></strong>- ¿Consumimos alimento de calidad? <br>- ¿Es rica en minerales el agua que bebemos?<br>- ¿Somos consciente del estrés que le generamos a nuestro cuerpo al intoxicarse con alcoholes, drogas, comida basura, etc?</li></ul><p>En el mundo tan acelerado en el que vivimos, no prestamos la debida atención a lo que consumimos y el efecto que esto puede acarrear en nuestra salud y desempeño. Una buena alimentación (calidad no cantidad) y beber suficiente agua trae más beneficios que cualquier rutina de ejercicios, píldoras milagrosas, o cualquier otra solución cortoplacista que nos seducen en estos tiempos.</p><h4>Dimensión mental</h4><p>- ¿Cómo cultivamos nuestra mente? <br>- ¿De que la nutrimos? <br>- ¿Estamos aprendiendo cosas nuevas?</p><p>Mucha gente gasta horas en revisar las fotos de nuestros amigos y familiares en las redes sociales, o viendo programas de televisión amarillistas, noticias desagradables, etc. Llenando nuestra mente de información inútil e incluso dañina. Nuestra mente así como nuestro físico, debe ser cuidada y cultivada, no solo evitando consumir información basura, sino ejercitando como un músculo; aprendiendo nuevas cosas. Hoy día en la era de la información, tenemos todo lo que necesitamos para aprender lo que sea desde la comodidad del hogar con solo acceso a internet, desde MOOCs, tutoriales, ebooks, etc. Podemos aprender hasta nuevos oficios si lo deseamos.</p><h4><strong>Dimensión espiritual</strong></h4><p>Ya estarán algunos leyendo esto con recelos y escepticismo, no intento venderles ninguna fe, religión o un Dios, sino, invitarles a pasar un momento con ustedes mismos, compartiendo un té, un café, una tarde de lluvia o cualquier momento que nos haga sentir bien, para tener una grata conversación con nosotros mismos y discutir acerca de nuestros valores, nuestras creencia, nuestra percepción de las cosas, sobre que nos da vida, que nos da sensación de bienestar, nuestros sueños, aciertos y errores. Acá un par de temas que puedes conversar contigo mismo:</p><p>- ¿Haces lo que te gusta? <br>- ¿Agradeces lo suficiente? <br>- ¿Cuáles son tus sueños? ¿Siguen vigente los viejos? ¿Qué estás haciendo para alcanzarlos?<br>- ¿Qué necesitas perdonar?</p><p>El autoconocimiento es una fortaleza importante que nos ayuda a avanzar y enfrentarnos a las limitaciones que se nos puedan presentar. Tomarse un par de minutos al día para esto te hará crecer de maneras increíbles personalmente.</p><h3>Ámbito Interpersonal</h3><p>La manera en la cual interactuamos y tratamos con otras personas, está muy ligada a cómo manejamos nuestras emociones y nos tratamos a nosotros mismos, por eso te invito a preguntarte:</p><p>- ¿Escuchas lo suficiente?<br>- ¿Sientes empatía por los demás?<br>- ¿Estás soportando alguna relación tóxica con pareja, amigos, compañeros de trabajo, etc? <br>- ¿Haces lo necesario para tener una buena actitud con los demás?<br>- ¿Tienes los amigos que desearías tener?</p><p>Los humanos somos animales sociales, la interacción con otros es un factor clave en nuestra supervivencia y en la calidad de vida que podamos tener, y la manera en que interactuamos es una proyección nuestra hacia el tercero; esto puede causarnos tantos problemas como oportunidades, todo depende de tanto nos preocupamos de esto.</p><h3>Ámbito profesional</h3><p>¿Les parece coincidencia que, la mayoría de las personas habilidosas y con éxito en sus carreras, llevan consigo una historia de arduo entrenamiento y/o estudios? Esto es un claro ejemplo de “Afilar el hacha”; antes de cortar me preocupo tener el hacha bien afilada, y en el proceso, si ya no corto como antes, entonces me detengo a afilarla. Esto puede pasar en cualquier carrera, proyecto, trabajo, sueños u otra cualquier cosa que emprendamos, siempre es importante preocuparnos por la preparación. Inclusos aquellas personas con dones necesitan preocuparse de afilar su hacha si quieren tener éxito, la habilidad sin compromiso y dedicación puede tener un valor casi nulo. Te invito a reflexionar sobre:</p><p>- ¿Te has sentido atascado en lo que haces?<br>- ¿No tengo los resultados que deseo, aun cuando me esfuerzo?<br>- ¿Me siento motivado en lo que hago?<br>- ¿Estoy repitiendo los mismos errores?<br>- ¿Me preparo lo suficiente?<br>- ¿Creo saber todo lo que necesito?<br>- ¿Cómo está las relaciones con mis compañeros? ¿Soy un buen compañero?<br>- ¿Estoy consciente de los espacios que tengo para mejorar?</p><p>Así como en las relaciones interpersonales, una debilidad que siempre nos juega en contra, y que es el mayor de nuestros enemigos es el ego, dejamos de aprender, escuchar, reflexionar incluso hasta de entender y comprender por ello. Todo el tiempo se nos presentan oportunidades para crecer, aprender, reforzar relaciones, crear nuevas, etc. Pero para ello es necesario saber que quiero, que hago para obtenerlo, qué tan consciente estoy de los errores que cometo, que cosas me faltan por aprender, de quien y/o de que, las oportunidades de mejoras que existen, etc. Una vez teniendo claro este tipo de cosas, ya tenemos todas las herramientas necesarias para entrar en acción e impulsar nuestras carreras y/o cualquier proyecto que queramos emprender.</p><h3>Un poco de historia</h3><p>Estoy proximo a cumplir un año como CTO en <a href="https://sosafeapp.com/">sosafeapp.com</a>, y ha sido una muy gratificante experiencia por:</p><ul><li>El equipo de trabajo.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*w-UFqEGbJG_lMnNRSe6y_Q.png" /></figure><ul><li>Lo que hacemos <a href="https://blog.sosafeapp.com/ayudamos-a-la-ciudadania-y-policia/">https://blog.sosafeapp.com/ayudamos-a-la-ciudadania-y-policia/</a></li><li>Las vidas que salvamos <a href="https://www.youtube.com/watch?v=BhznKORNCUQ">https://www.youtube.com/watch?v=BhznKORNCUQ</a></li><li>la mejora en la calidad de vida de nuestra comunidad <a href="https://blog.sosafeapp.com/atrapamos-a-un-narcotraficante/">https://blog.sosafeapp.com/atrapamos-a-un-narcotraficante</a></li><li>La gratitud de nuestros usuarios</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*P5Zlq1LgGA5YQ0vQGPDFhg.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vaCfjRNTT2k8XFNy0OJQuw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WHhUa43oKWfhjme5Jxf8oQ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*SSCvsCo_mF5ptvt--jaTeA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*98QU2lzspOtvLDU31ibZiQ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qJicoanmrGJRCDGnqN7NJg.png" /></figure><p>y sobre todo, por el gran crecimiento que hemos tenido como organización al emplear este tipo de hábitos.</p><p>Hace un año, la cantidad de funcionalidades que se publicaban no llegaba a ser una al mes, la mayoría del trabajo empleado era para corregir errores y hacer trabajos manuales todo el tiempo, lo que hacía casi imposible el poder trabajar en cosas nuevas, y en las que se podían por temas de tiempo y compromisos, se realizaban de maneras poco eficiente generando deudas técnicas que a su vez, se transformaban en más errores para preocuparse en el futuro y este escenario se había ya establecido desde hace mucho tiempo.</p><p>En la primeras semanas, mi pregunta fue como la del leñador jefe <em>“¿Cuándo fue la última vez que afilaste tu hacha?” </em>Pero está dirigida hacia toda la organización. Poco a poco recopilando la suficiente información y con el compromiso de todos, comenzamos a hacer una lista priorizada de las partes del hacha que debíamos afilar. Hoy día puedo decirles que con esto hemos conseguido:</p><ul><li>Aumentar 8 veces la escalabilidad de nuestros servicios (calculado en poder de cómputo/rendimiento).</li><li>Reducir a la mitad los costos de infraestructura.</li><li>Agregar tests en nuestros desarrollos (no existían tests).</li><li>Reducir considerablemente la responsabilidad de nuestro servicio perteneciente a un gigante y grotesco monolito, separados en diferentes microservicios.</li><li>Reducir el trabajo manual diario a la mitad con herramientas internas.</li><li>Aumentar considerablemente los pasos a producción en los distintos servicios y aplicaciones.</li><li>Ganar 5 veces velocidad en desarrollo, consiguiendo experimentar rápido en nuevas funcionalidades, mejoras y hasta ideas propuesta por el equipo (calculado en semanas promedios para desarrollar un feature nuevo).</li><li>Expandirnos a más países, el más nuevo Ecuador <a href="https://blog.sosafeapp.com/bienvenido-ecuador/">https://blog.sosafeapp.com/bienvenido-ecuador/</a></li><li>Construir un equipo humano excepcional.</li></ul><p>Cuando la rutina diaria, la sobreinformación, tu estado de ánimo o cualquier cosa que influya de manera negativa en tu vida y entorpezca tu desempeño y el camino a alcanzar lo que quieres, recuerda darte un tiempo para <strong>afilar el hacha. </strong>No seas como el leñador joven y respondas <em>“No tengo tiempo para ello, he estado muy ocupado cortando árboles”.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f5fe5be7dfb6" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Machine Learning Ideas: Separating background jobs with machine learning]]></title>
            <link>https://medium.com/lemontech-engineering/machine-learning-ideas-separating-background-jobs-with-machine-learning-12a0ad641f0e?source=rss-64c23a1254a7------2</link>
            <guid isPermaLink="false">https://medium.com/p/12a0ad641f0e</guid>
            <category><![CDATA[unsupervised-learning]]></category>
            <category><![CDATA[ideas]]></category>
            <category><![CDATA[services]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[machine-learning]]></category>
            <dc:creator><![CDATA[Javier Segovia]]></dc:creator>
            <pubDate>Mon, 20 Nov 2017 13:21:40 GMT</pubDate>
            <atom:updated>2018-12-12T13:16:02.814Z</atom:updated>
            <content:encoded><![CDATA[<p>In computer programs it is common to trigger/schedule tasks that should be performed without user intervention. These kinds of tasks could be: logging, system monitoring, scheduling, user notification, etc. These tasks are not necessary to be executed immediately and for some cases, the end-user will never notice them, and for that reason, they should be executed separately from the current thread.</p><p>A background or asynchronous job (or task) is one that is processed outside of the usual request/response workflow that is part of any modern web framework. Normally, web applications receive a request from the outside world, do some processing (such as querying a database) and immediately return a response within a few milliseconds. This is the normal pattern that we have all become accustomed to when developing applications for the web, and is known as synchronous communication.</p><p>Asynchronous tasks on the other hand, are those that may be started from a normal web request, but require a longer time to complete than the normal request. Because these requests cannot be processed immediately and return a response, they are known as asynchronous. In order to avoid interrupting the normal synchronous workflow of an application, asynchronous tasks are normally processed on a separate thread or are spawned as a separate process entirely.</p><p>These asynchronous jobs are added to job queues, that are data structures maintained by a job scheduler software containing jobs to run, so the goal of this project is to separate the jobs into different job queues, grouping them by similarity, helping us to use wisely our server’s resources assigning queues to different machines that fit with the computational requirement for each job.</p><h3>Problem Statement</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*sJv0nUaHlZOOtk3r8Df8NA.png" /></figure><p>We were having difficulties escalating our background jobs because we have to process a lot of tasks that are similar categorically, but each one of them can use external services, do heavy processes to accomplish the task, the service requested can be allocated in different countries, among other things that can make it unique or similar to other tasks. In addition to the above, our customers are not subscribed to every task, so that means customers that have tasks types that don’t consume a lot of resources and are really fast to complete, have to be in the same queue with those tasks that take too long to be performed.</p><h3>How did we solve it?</h3><p>Simple, we replicated our server until we could complete our tasks in N time.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/1*If8i4Kfa_1bq1ymnciFbWA.png" /></figure><p>This solution is not good enough for two main reasons:</p><ul><li><strong>It’s not optimal:</strong> We have to deploy our services in machines with the same specs, even when some of the tasks need none computational power and others do.</li><li><strong>It’s not fair: </strong>The customers who are subscribed to fewer tasks than others, and who’s tasks belongs to a group who have lower complexity, will be in the same queue with the rest and diverse tasks.</li></ul><p>Because of that, we started to design a solution to get the best result.</p><h3><strong>First Approach: Original Solution</strong></h3><p>The goal of the solution is to cluster these tasks to serve them into different queues with specific features that fit perfectly to solve the problems in order to escalate efficiently.<br>As inputs, we are given information such as customer reference, type of task, where the service requested by the task is allocated, if the task has to perform an external process, if the task has to perform a heavy process, the count of this kind of task for each customer, and the average duration to complete the task. Our original solution was to separate the tasks by different criteria:<br>• Split the tasks by country.<br>• Split every subgroup by use of external processes.<br>• We observed that Task #11 from Chile had too many jobs to process, which led us to decide that this task should be separate from the rest.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/679/1*1U4ySt37DvKk_Qu1IrFlww.png" /></figure><p>This solution was developed after analyzing our clients and data, and understanding what we had and how our service was behaving. We realized this solution could work perfectly at the moment but, because we are constantly adding new tasks and having more customers, in the near future this solution would be obsolete and we would have to repeat the whole analyzing job again in order to develop another solution that fits our current data.</p><h3><strong>Final Solution: Machine Learning</strong></h3><h4>Data Exploration</h4><p>As part of this problem, I’ve collected all tasks data (178571 records) grouped by customers, and this dataset contains 1503 records, each record is represented by a row in the csv file, and has several features given for it. I have described each feature given briefly below:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wlSx4JszxzWE2iyGNUT7jw.png" /></figure><p>This means we have a total of 6/7 features that we can train on, which includes 2 categorical, 2 boolean and 2 numerical values.</p><h4>Data Visualization</h4><p>After creating A LOT of plots and graphs, I got a better understanding about what we had, and how each task is distributed and determined where the limits for each one is, and we can conclude that:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*01fICYL5P7hu4J1y1jBpZw.png" /></figure><h4>Algorithms and Techniques</h4><p>The algorithm I used to create the clusters was <a href="https://en.wikipedia.org/wiki/K-means_clustering"><strong>KMeans clustering</strong></a>; this is a method of vector quantization that is popular for cluster analysis in data mining, and it works partitioning the input points (observation) into K distinct sets (clusters). The concept of this algorithm is as follows:</p><p>For the K clusters, select the initial K-centroid, this usually is chosen randomly, then for these initial centroids, the algorithm proceeds by repeating these two main steps:</p><p><strong>1. Cluster Assignment: </strong>Each observation (each point in the dataset) is assigned to a cluster centroid such as the Within-Cluster Sum of Square (WCSS) objective function is minimized. This can often be translated to assigning each observation to the closest cluster centroid, which minimizes the WCSS for many distance metrics.</p><p><strong>2. Update Centroids: </strong>After all of the observations have been assigned to a cluster centroid, each centroid is recomputed. For each cluster, the new centroid is calculated by averaging the observations that were assigned to it, which means computing the ‘mean’ of the observations.</p><p>These steps repeat until the algorithm ‘converges’, this is detected by several ways, but the most common is to run until none of the observations change cluster membership.</p><p>I’ve chosen this algorithm because it is easy to understand, easily adaptable, works well on small or large datasets, it’s fast, efficient and has a good performance; on the other hand, it’s cons is that we need to choose the proper number of clusters, which can be solved through several techniques, but since it is a NP-Hard algorithm it’s not recommended if you are looking for performance with a lot of data. For this problem, after the preprocessing, the data’s length won’t be an issue, therefore the con doesn’t apply.</p><h4>Data Preprocessing</h4><p>To convert the data into a format which our model can effectively use, as only numerical features are supported, the feature “country” must be transformed and the whole data should be grouped by task.</p><p>To group the data by tasks, I created different datasets to merge them later, and these datasets are:</p><p>• <strong>Task’s count:</strong> Sum of each count per task type<br>• <strong>Average Duration:</strong> Group and calculate the mean for the avg_duration of each type of task.<br>• <strong>Discrete values:</strong> Unique values for the combination of the discrete values (Task, Country, External Process, Heavy Process).</p><p>Then I did some feature engineering to create new features:</p><p>• <strong>Customer by task:</strong> How many customer belongs for each task.<br>• <strong>Total Duration:</strong> Multiplication of each task’s count and average duration. It represents how long it takes to complete all tasks per type.</p><p>Then all this datasets can merge in a single one.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rysBlljKTRW3oaJDkVcJ_Q.png" /></figure><p>Because I created the new feature Total Duration which is the result of a multiplication of vectors, I decided to drop it’s multipliers (count and avg_duration) to keep just one feature of that kind instead of 3.</p><p>Because country’s values are string, I needed to transform them into numerical values, then I converted them into dummy variables.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*lR6APfzcAPbx0UDK9wBUYQ.png" /></figure><p>In the data above, we can observe how distant are some column values to another, like the boolean ones and the continuous ones. This will affect the calculation to play down some features’ importance, so we need to scale them using a <a href="http://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.StandardScaler.html"><strong>Standard Scaler</strong></a>, that standardizes features by removing the mean and scaling to unit variance. The image below will show a sample of the data scaled.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jyWPTwtUVVuDYouOTEQYEw.png" /></figure><p>For the algorithm I’ve chosen to provide a solution to this problem, it depends for distance measure, points near each other are in the same cluster; points far apart are in different clusters. But in high dimensional spaces, distance measures do not work very well, because of that I reduced the dimension of the data, applying Principal Component Analysis (<a href="https://en.wikipedia.org/wiki/Principal_component_analysis"><strong>PCA</strong></a>), selecting the most useful components by their explained variance.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rtFLHxQxfAwY0-ewsBeJOQ.png" /></figure><p>As we can see the first 6 component are the most useful, but I picked pick the first 4 to make it shorter, before reducing the dimensionality, our data’s shape was (26, 35) and now is (26, 4).</p><h3>Results</h3><p>Before to analyze the results, let’s make a recap about our original solution<br>Our original solution was separate the task for different criteria:</p><p>• Split the tasks by country.<br>• Split every subgroup by use of external processes.<br>• We observed that Task #11 from Chile had too many jobs to process, which led us to decide that this task should be separate from the rest.</p><p>This solution was developed after analyzing our service by clients and data, understanding what we had and how is behaving. Our original solution is similar to our ultimate solution, let’s compare the results:</p><p>Comparing the results with our benchmark model (our human original solution) we can see that, each cluster contain only one country, the Task #11 is alone and those tasks that use external process are separated from the rest as we expected, except for Colombia’s tasks that are separated by if they perform heavy process, but it’s OK because that feature wasn’t included in the original solution.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/555/1*KTc9TorPfbNEkBry9GJ-Jg.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/551/1*IUffup3OQMlt7SAS9S5YEg.png" /></figure><p>Comparing the results with random data :</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/797/1*dwv5DK-42qdDIcz4Yry4kA.png" /></figure><p>We can observe that our model is robust enough, it groups almost all sample data in their corresponded cluster as our final model will do, except the Task #25 in cluster 8, it doesn’t make sense, I’d expect it to be assigned to cluster 3 (because the country) or alone as cluster 4 and 9 because it use a heavy and external process.</p><h3>Conclusion</h3><p>This solution is about to create an unsupervised model to clustering the data and to separate job queues by clusters, to achieve that, first I had to create a pipeline to preprocess the data, convert all non-numeric values to numerics, group the data by task’s types, do some feature engineering and remove unnecessary features. Then train the model to assign a cluster to each task.</p><p>This solution will automatized the analysis work that our team had to do to understand our process for then, select the best way to separate the tasks in different job queues, saving us a lot of time.</p><p>Machine learning is a method of data analysis that automates analytical model building, and it can be used for anything we can imagine, but it will be useful to develop solutions using algorithms that iteratively learn from data, machine learning allows computers to find hidden insights without being explicitly programmed where to look.</p><p>If you have enough data, give a try to build a model instead of coding complex algorithm solutions that should be tuned passed the time, and have fun with it!.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=12a0ad641f0e" width="1" height="1" alt=""><hr><p><a href="https://medium.com/lemontech-engineering/machine-learning-ideas-separating-background-jobs-with-machine-learning-12a0ad641f0e">Machine Learning Ideas: Separating background jobs with machine learning</a> was originally published in <a href="https://medium.com/lemontech-engineering">lemontech-engineering</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Cazando bots en Twitter]]></title>
            <link>https://medium.com/5rabbits-engineering/cazando-bots-en-twitter-f173e2315185?source=rss-64c23a1254a7------2</link>
            <guid isPermaLink="false">https://medium.com/p/f173e2315185</guid>
            <category><![CDATA[python]]></category>
            <category><![CDATA[bots]]></category>
            <category><![CDATA[twitter]]></category>
            <category><![CDATA[politics]]></category>
            <category><![CDATA[venezuela]]></category>
            <dc:creator><![CDATA[Javier Segovia]]></dc:creator>
            <pubDate>Thu, 08 Jun 2017 19:03:19 GMT</pubDate>
            <atom:updated>2017-08-16T16:06:08.573Z</atom:updated>
            <content:encoded><![CDATA[<p>Imagínense vivir en un país en dictadura, donde el dia anterior había sido el <strong>Dia del Trabajador</strong> y el salario mínimo mensual no llega a 50 USD, paralelamente habían<a href="https://www.google.com/url?sa=t&amp;rct=j&amp;q=&amp;esrc=s&amp;source=web&amp;cd=2&amp;cad=rja&amp;uact=8&amp;ved=0ahUKEwjuzN-5wdLTAhWBipAKHbLiCEAQqUMIKjAB&amp;url=http%3A%2F%2Fwww.elnuevoherald.com%2Fnoticias%2Fmundo%2Famerica-latina%2Fvenezuela-es%2Farticle147825589.html&amp;usg=AFQjCNGPwgH2m6eH_rCQyci0OcLHop8qZA&amp;sig2=dNafQ8b9hCpHrLBzPl9T4w"> disturbios y manifestaciones</a> en todo el país (porque el país sufre<a href="https://www.google.com/url?sa=t&amp;rct=j&amp;q=&amp;esrc=s&amp;source=web&amp;cd=8&amp;ved=0ahUKEwjW5vjGwdLTAhVEQ5AKHcahA60QFghKMAc&amp;url=http%3A%2F%2Fwww.eltiempo.com%2Fnoticias%2Fcrisis-en-venezuela&amp;usg=AFQjCNGffjGkccUKANNdLkaG_Ae8jw3MSA&amp;sig2=VHisGCalKYeFI2FK1fOI9Q&amp;cad=rja"> crisis económicas, sociales, de salud, etc.</a>) siendo reprimidas de forma violenta por el poder que juro proteger al ciudadano, un dia terrible, pues ese es mi país de origen, mi amada Venezuela.</p><p>Para mejorar la situación, el presidente convoca a una <a href="https://www.google.com/url?sa=t&amp;rct=j&amp;q=&amp;esrc=s&amp;source=web&amp;cd=6&amp;cad=rja&amp;uact=8&amp;ved=0ahUKEwi8o4jvwdLTAhVEQ5AKHcahA60QFghCMAU&amp;url=http%3A%2F%2Fwww.el-nacional.com%2Fnoticias%2Fgobierno%2Fmaduro-convoco-una-constituyente-comunal_179943&amp;usg=AFQjCNHBvlVf-h3shtusQ2SDmAipnTsXfw&amp;sig2=8GL5l9W82DffVG2yPzZyoA">“Constituyente Comunal”</a> el cual genera todo tipo de repudio nacional como internacional, aun así al siguiente dia en el <strong>twitter</strong> en medio de protestas y repudios la etiqueta pro-gobierno <strong>#AConstituyentePorLaPaz</strong> se mantiene posicionada, este fenómeno ocurre todos los días, al tener todos los medios censurados en el país, todo es color de rosa, pero muy contrario a la versión de las redes sociales, y al no poder censurarla, tratan de manipularla desvirtuando la opinión publica teniendo <strong>trending topics</strong> pro gobierno, para así tener una prueba de que el “pueblo” (a.k.a <strong>bots</strong>) están con ellos, este es un secreto a voces, muchas personas han demostrado con fotos, videos, etc como operan los bots, pero yo aproveche el fin de semana largo, y me dedique a demostrarlo como mejor lo se hacer, creando <strong>bots</strong> para detectar <strong>bots</strong> :D.</p><p>Cabe destacar que esto no es solamente para analizar los bots del gobierno, hay otros personajes que se dedican también a posicionar etiquetas de formas fraudulentas, sobre todo en las noches, así que parte de lo que mostrare no funcionara contra los <strong>bots</strong> <strong>“chavistas”</strong> porque usan estrategias distintas.</p><p>Así como este post lo hago en español, no lo haré tan técnico como mis posts anteriores, mostrare simplemente los resultados obtenidos al analizar para ser exacto <strong>15563 tweets</strong> con la etiqueta <strong>#AConstituyentePorLaPaz</strong></p><blockquote>Aquellos que son programadores, esto lo hice usando <a href="https://github.com/tweepy/tweepy">https://github.com/tweepy/tweepy,</a> y <a href="http://pandas.pydata.org/">http://pandas.pydata.org/</a>, luego subiré el código fuente</blockquote><p>Para detectar bots me base en varios criterios</p><ul><li>Tweets originales (no retweets) duplicados por distintas cuentas, es muy sospechoso que varias personas piensen igual (aunque eso es parte del modelo socialista, el pensamiento único :D)</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OBwULKspP4a4WHVEN4Zsfw.png" /></figure><ul><li>Tweets falsos, como les mencione anteriormente, con los otros personajes que posicionan etiquetas, tienen varias cuentas que publican varios tweets con números o palabras aleatorias que no tienen que ver con la etiqueta.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yJ6Oi4pXE-oEZPK_bKp0UQ.png" /></figure><ul><li>Tweets masivos por el mismo usuario, hay varias cuentas que se dedican a twitter y a dar RT masivamente a los tweets con estas etiquetas</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/326/1*Ah8xA47gIFlwJzcuxMIHVw.png" /></figure><ul><li>Tweets provenientes de herramientas de automatización, algo que encontré muy resaltate, era como estos bots operaban mediante el uso de TweetDeck, Botize y IFTTT</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3bRzCCeW2_XzUrl7MH_qug.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*KNOFHpZL1cQS5SRQdr_vSw.png" /></figure><ul><li>Y por ultimo, el mas descarado, RTs hecho masiva y simultáneamente, esta comparación la hice buscando todos los tweets hecho rt al mismo tiempo, es decir, con hora, minuto, segundos y mili-segundos!, sin dejar abierto la posibilidad de margen de error</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*U-ye_kS9ZSgE752Y-SpGwA.png" /></figure><p>En resumen, si bien despreciamos todas las pruebas excepto la ultima de los retweets simultáneos, representa mas del 15% de los tweets, por la densidad de los mismos, posicionan de manera eficaz las etiquetas, el resto de los tweets y menciones si se generan de manera orgánica, ya que algunos las respaldan y otros simplemente la condenan, de ambas formas mantienen la posición de la etiqueta, pero solo necesitaba el primer empujón, gracias al pueblo (bots)!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f173e2315185" width="1" height="1" alt=""><hr><p><a href="https://medium.com/5rabbits-engineering/cazando-bots-en-twitter-f173e2315185">Cazando bots en Twitter</a> was originally published in <a href="https://medium.com/5rabbits-engineering">Crafting products with joy</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>