<?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 Carlos Justiniano on Medium]]></title>
        <description><![CDATA[Stories by Carlos Justiniano on Medium]]></description>
        <link>https://medium.com/@cjus?source=rss-448274adae5e------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*BZSVzezWGU7dOaciAUWA_w.png</url>
            <title>Stories by Carlos Justiniano on Medium</title>
            <link>https://medium.com/@cjus?source=rss-448274adae5e------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 07:47:53 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@cjus/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[A Daoist Hacker Manifesto]]></title>
            <link>https://medium.com/daoisthacker/a-daoist-hacker-manifesto-66e91c3be794?source=rss-448274adae5e------2</link>
            <guid isPermaLink="false">https://medium.com/p/66e91c3be794</guid>
            <category><![CDATA[taoism]]></category>
            <category><![CDATA[lifehacks]]></category>
            <category><![CDATA[daoist]]></category>
            <category><![CDATA[taoist]]></category>
            <category><![CDATA[daoism]]></category>
            <dc:creator><![CDATA[Carlos Justiniano]]></dc:creator>
            <pubDate>Sat, 22 Mar 2025 14:05:49 GMT</pubDate>
            <atom:updated>2025-03-22T22:52:17.116Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*eYNY1IFfj7sD0QePlteF6w.png" /></figure><p>First published on March 22, 2025 <a href="https://creativecommons.org/licenses/by-nc-nd/4.0/">Released under the Creative Commons License CC BY-NC-ND 4.0</a> — co-published on DaoistHacks.com</p><p><em>A guide for those of us who seek to transcend limitations, harmonize with the flow, and reshape reality through the mastery of systems.</em></p><p>All things — computers, networks, energy, life — follow patterns. To hack is to see beyond appearances and understand the underlying flows. The skilled hacker does not fight the system but moves with it, bending it effortlessly to their will.</p><p>As the young monk in The Matrix said:</p><blockquote><em>“Do not try to bend the spoon — that’s impossible. Instead, only realize the truth… there is no spoon. Then you will see it is not the spoon that bends, it is only yourself.”</em></blockquote><p>A hacker knows that limits are illusions, that systems are only as rigid as our perception of them. When we stop resisting and start understanding, we can reshape anything — not by force, but by moving with the Dao, the underlying code of reality.</p><h3>What it means to be a Daoist Hacker</h3><p>A Daoist Hacker is a master of systems — whether digital, physical, or energetic — who harmonizes with the natural flow of things to create effortless transformation. They see reality as a programmable code or a living Dao, understanding that true power comes not from force but from alignment with deeper principles.</p><p>Rather than fighting obstacles, a Daoist Hacker flows like water, adapting, learning, and reshaping reality with minimal effort. They study hidden patterns — whether in computer networks, nature, or human consciousness — and find elegant, simple ways to rewrite the rules.</p><p>They explore, tinker, and innovate, but they do so with awareness and balance, respecting the harmony of systems rather than causing unnecessary disruption. They believe knowledge should be free, systems should be open, and that intention can shape outcomes.</p><p>Their approach blends Wu Wei (effortless action) with deep focus, allowing them to unlock doors that others think are closed — whether in cyberspace, personal growth, or society itself.</p><p>Hackers doesn’t seek control but understanding. They cultivate mastery, not ego. They recognize that everything — whether a line of code or a human mind — can be optimized, reprogrammed, or transformed through the right action at the right time.</p><p>Reality is a system. The Dao is the source code. A hacker learns to read, rewrite, and flow with both.</p><h3>Wu Wei: The Effortless Hack</h3><p>The hacker does not force change but finds the point where the system is ready to shift. With the lightest touch, a locked door opens, a network bends, or a mind awakens. True skill is not brute force but harmony with the nature of things.</p><h3>Information Wants to Flow</h3><p>The Dao flows freely, and so must knowledge. To hoard understanding is to dam a river — eventually, it will break free. Share wisdom, teach those who seek, and let information move like water, filling all spaces naturally.</p><h3>The Mind Shapes Reality</h3><p>The weak hacker sees obstacles; the master sees opportunities. The hacker knows that intention shapes the code of the universe — whether digital or energetic. What you focus on expands. A single insight can shift an entire system.</p><h3>Be Like Water, Adapt and Overcome</h3><p>When faced with a barrier, flow around it. When blocked, seep into the smallest cracks. If contained, rise and reshape the environment. Water does not resist; it changes form. The hacker who understands this is unstoppable.</p><h3>Respect the Balance</h3><p>Every system has yin and yang, light and dark, structure and chaos. A wise hacker does not seek change for its own sake but moves in balance. A small disruption can lead to great harmony when applied at the right moment.</p><h3>The Path is Mastery, Not Ego</h3><p>A true hacker is a student of the Dao, forever learning, forever refining their craft. To seek status, fame, or control is to stray from the way. The greatest power is not seen — it moves in silence, known only by its effects.</p><h3>The Universe is Open-Source</h3><p>The cosmos is an unfolding code, freely available to those who look deeply. Study the patterns, learn the flows, and rewrite what needs rewriting. The Dao is not fixed, and neither is the system.</p><h3>Simplicity is the Ultimate Hack</h3><p>A thousand lines of code can be replaced by one. A complex system can be undone by a single point of leverage. The Dao is simple, yet it contains everything. The skilled hacker seeks elegance, reducing effort to the absolute minimum.</p><h3>Let Go, and Let the Code Run</h3><p>The greatest hack is to know when to stop. The Dao moves on its own; no need to force it. Once the script is written, let it execute. Once the flow is in motion, step back and observe. In letting go, we master the greatest system of all.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=66e91c3be794" width="1" height="1" alt=""><hr><p><a href="https://medium.com/daoisthacker/a-daoist-hacker-manifesto-66e91c3be794">A Daoist Hacker Manifesto</a> was originally published in <a href="https://medium.com/daoisthacker">DaoistHacker</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Cloud-based file transfer at scale]]></title>
            <link>https://medium.com/@cjus/cloud-based-file-transfer-at-scale-a32ceef68fc5?source=rss-448274adae5e------2</link>
            <guid isPermaLink="false">https://medium.com/p/a32ceef68fc5</guid>
            <category><![CDATA[serverless]]></category>
            <category><![CDATA[microservices]]></category>
            <category><![CDATA[nodejs]]></category>
            <category><![CDATA[cloud-computing]]></category>
            <category><![CDATA[aws-lambda]]></category>
            <dc:creator><![CDATA[Carlos Justiniano]]></dc:creator>
            <pubDate>Fri, 12 Aug 2022 02:33:39 GMT</pubDate>
            <atom:updated>2022-08-12T02:33:39.100Z</atom:updated>
            <content:encoded><![CDATA[<p>Using NodeJS, Redis, Docker and AWS Lambda</p><blockquote>Note: This article was first published at <a href="https://medium.com/hackernoon/cloud-based-file-transfer-at-scale-63d8e2dacb3a">HackerNoon on October 2018</a>. Since that time HackerNoon has gone its separate way the hosted article on Medium looks horrible against the bright green background. So I’ve copied the article here in its unedited entirety.</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3UMUhuabDE54hcmbS-R8qg.jpeg" /></figure><p>Recently, I had an opportunity to build a system for transferring large numbers of files from one cloud Content Delivery Network onto another. This post describes the key components and the overall architecture.</p><p>The file content in question was nothing less than our entire video library for <a href="https://shop.flywheelsports.com/">FlyAnywhere</a> — the live and video on-demand fitness service we created at <a href="https://www.flywheelsports.com/">Flywheel Sports</a>.</p><p>During the past year, we’ve amassed a sizable video library. Copying that library from one CDN to another would be a time-consuming operation — but one that would eventually complete — even if it took a few days to transfer. However, more important than a one time transfer is the fact that an ideal solution would need to continually transfer content after each newly recorded class. This is where transfer time equates to fresh content availability.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CjnsMYobxZu5b6nVZ-nZQA.png" /><figcaption>A Live broadcast at our FlyLive Studio in NYC</figcaption></figure><p>I wish I could say that the transfer operation was simply a matter of invoking an SCP/SFTP file transfer. But there was no such luck. Files in the primary CDN are stored in an <a href="https://en.wikipedia.org/wiki/Object_storage">object storage container</a> (think S3) and there’s no direct access to a directory structure per se — just a flat bucket with path-like filenames.</p><p>Additionally, each video is stored as an <a href="https://en.wikipedia.org/wiki/HTTP_Live_Streaming">HLS</a> stream — a live streaming format consisting of multiple manifest files and lots of individual audio and video segments stored as files. A typical 30-minute video can consist of thousands of segments due to the number of multi-channel video streams required to support <a href="https://bitmovin.com/adaptive-streaming/">adaptive streaming</a>.</p><p>So the task at hand required reliably transferring an entire collection of videos with millions of individual file segments. Terabytes of data, in essence, a highly-scalable file sync operation.</p><p>Naturally, it’s the “reliably transferring” phrase above that requires special attention. File transfers fail due to broken and or timed out network connections. A blind copy just won’t do. Each file needs to be properly tracked and accounted for.</p><h4>Enter Lambdas</h4><p>The source files that needed to be copied are already in the cloud and the destination is <a href="https://hackernoon.com/tagged/aws">AWS</a> S3 for use with CloudFront. Both are ideal scenarios for servers which are also already in the cloud. My first consideration was simply to use a Docker Swarm cluster to manage a collection of microservices. Such an approach would rely on multiple container replicas to ensure distribution across available machine CPU cores. Then I thought, “<em>Why not try lambdas?</em>” After all, file transfer tasks are what’s known in computer science as an <a href="https://en.wikipedia.org/wiki/Embarrassingly_parallel">embarrassingly parallel</a> problem — given that each file can be individually transferred.</p><p>AWS <a href="https://aws.amazon.com/serverless/videos/video-lambda-intro/">Lambda</a> seemed like an ideal solution to perform the transfer at scale. Lambda’s offer scalable compute resources without the need for server provisioning and maintenance. Enter the so-called server-less revolution.</p><p>Quickly thereafter the <a href="https://en.wikipedia.org/wiki/Dopamine">Dopamine</a> rush hit… <em>whaaaat?</em> <em>I could have thousands of machines simultaneously moving files? Sign-me-up!</em></p><h4>Introducing FLAnts</h4><p>Using Lambdas, the process behaves like a swarm of ants carrying objects back to an ant colony. And given that the files in question originate from our Flywheel Live production studio — I decided on the name FLAnts (FlyLive Ants). Cheesy — I know! Fortunately, I’m not in marketing…</p><p>FLAnts consists of a command line tool and three microservices. Each one built using <a href="https://hackernoon.com/tagged/nodejs">NodeJS</a>. The fifth module is the lambda code — also written in Javascript for NodeJS. The microservices run in a Docker Swarm cluster.</p><p>Let’s take a closer look at the underlying process. The first order of business required determining which content needed to be copied. A process called the class-scanner pulls class metadata from our cloud MongoDB and performs basic validation on each class. Because not all classes stored in the database are actually broadcasted. Some are test classes, others are aborted classes — imagine a director yelling <em>“cut!”</em>.</p><p>For every validated class, the class-scanner pushes a job onto a Redis job queue.</p><p>A second process is the class-crawler. The process gets its name from the similar process of <a href="https://en.wikipedia.org/wiki/Web_crawler">web crawling</a>, an underpinning of Internet search engines. The class-crawler, pops a job from the Redis queue and performs the task of reading each classes HLS manifest file and crawling it to collect a list of individual HLS segment file names. Each segment’s metadata is used to populate another job queue which identifies each file that needs to be copied. Each resulting job can be distributed to one of thousands of Lambda instances. It’s also worth noting that the class-crawler doesn’t have to be a single process. On a large scale multi-core machine we fire off one per CPU core. And each pulls atomically from the job queue. Amazon has sports car-class machines featuring 72 and 128 cores. It is a great time to be in tech!</p><p>A third process called, segment-transfer, pops transfer jobs from the Redis queue and dispatches them to lambda instances. The role of the segment-transfer process is to act as an orchestrator or traffic cop of sorts. Also like the class-crawler, the segment-transfer process can run as multiple processes for all available CPU cores — across multiple machines.</p><p>Our forth and last process is called the aggregator. Its role is to process completed jobs and assemble them based on their respective classes. When all of the segments of a class have been processed, the transfer operation is deemed successful.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7XfNxl2cDSzRFBFna2muDA.png" /></figure><p>The diagram above shows the four process types which can either run on a large multi-core machine or across a number of EC2 instances. Each connects to a Redis instance for job queuing. The segment-transfer process is shown dispatching lambdas which in turn copy data from one CDN to another.</p><p>The reason for this overall approach is that any one of the four process types can fail and require a restart. That was especially true during the development phase and is a good example of <em>designing software with failure in mind</em>. That’s something that’s vital when building distributed systems.</p><h4>Message Queuing FTW</h4><p>As described above, all four processes have a very important feature in common: each relies extensively on message queuing. The class-scanner queues messages, the class-crawler, and segment-transfer both dequeue and queue messages. And the aggregator dequeues messages and generates a transfer report.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/764/1*p9VWGZOlbZwlirShHpxF-g.png" /></figure><p>A key reason for using message queuing is to control the flow of work between processes. Each process is blazing fast and it’s not uncommon to max CPU usage and blow through both memory and socket handles. It’s far better to queue messages and allow processes to consume them as fast as they can — but not faster!</p><p>Another significant benefit is that queuing allows for multiple instances of a particular task to participate in the message processing.</p><p>Additionally, queuing and the use of multiple process instances allows NodeJS-based applications to take full advantage of multicore machines.</p><p>Queues also assist in decoupling applications. In the illustration above each service type can have instances on the same and across networked machines; so long as the queuing service is accessible.</p><p>I chose to implement my processes using NodeJS and<a href="https://www.npmjs.com/package/hydra"> Hydra</a>. Hydra has built-in message queuing which sits on top of Redis. So implementing the above queuing couldn’t be easier. Here’s an example from the segment-transfer process.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/48cd7391b47cfc826eab17ccdde7c8fb/href">https://medium.com/media/48cd7391b47cfc826eab17ccdde7c8fb/href</a></iframe><p>I only need to specify which service receives the job (using the to field) and include a bdy (body) object which contains the actual job details.</p><p>The aggregator service only needs to dequeue messages when it’s ready:</p><p>let message = await this.hydra.getQueuedMessage(&#39;aggregator-svcs&#39;);</p><p>The aggregator is responsible for dequeuing jobs messages which contain the result of each lambda instance’s file transfer.</p><p>I used the queuing features built into Hydra, but you don’t need to use Hydra to do something similar. See this <a href="https://gist.github.com/cjus/719d77dd9816115d94083234c293c18b">sample gist</a> showing the relevant Redis commands.</p><h4>Batching</h4><p>The segment-transfer service reads transfer jobs from a queue and dispatches them to lambda instances. Also, keep in mind that there are multiple segment-transfer instances (replicas) running.</p><p>Rather than attempt a massive lambda launch, each segment-transfer service reads 100 jobs and organizes them into a single batch. That results in 800 lambda invocations on an eight core machine with eight segment-transfer instances. That works out well given that by default there’s a maximum of 1000 concurrent lambda invocations. In practice I haven’t seen more than 500 simultaneous invocations due to the overhead of processing jobs into batches and launching lambda instances. Running twice the number of segment-transfer replicas per core would change that.</p><h4>Error handling</h4><p>Occasionally, a lambda instance will fail to perform a transfer — shocking I know! When that happens the issue can be attributed to a network socket break or timeout. However, this isn’t an issue, because the entire system relies on messaging queuing so it’s just a matter of re-queuing the failed job.</p><p>Each job message contains a retry field which starts at three. Each failed attempt reduces the count by one and should the count reach zero then the job is marked as a complete failure. This has never happened. The most I’ve seen is a single retry resulting in success.</p><p>This all works exceedingly well. However, re-queuing messages causes them to get pushed to the end of their job queues. So what ends up happening at run-time is that jobs don’t complete until much later in the process. I noticed this behavior when I added a web-based visualizer to the aggregation service.</p><p>The classical solution is to use a priority-based queue. A priority queue would have allowed me to re-queue jobs with a higher priority to ensure those jobs didn’t have to move the end of the line. Redis support this <a href="http://charlesnagy.info/it/python/priority-queue-in-redis-aka-zpop">using the zpop</a> operation, which is something I may later revisit. Yet another potential solution is to maintain a separate job queue for failed operations and pull from that queue before pulling from the main job queue. It’s a code complexity trade-off.</p><p>Instead, I simply maintained an in-process re-queue list and pushed those jobs onto the next batch operation.</p><h4>Visualizing the file transfer</h4><p>To better see the actual behavior of the overall system I decided to build a web-based visualizer. The thought of rendering each of those virtual ants (I mean lambdas) as they transferred file segments — definitely warranted a side-hack.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/748/1*AUSPnLtpUj61DAbM7MTdUQ.gif" /></figure><p>I added a dashboard endpoint to the aggregator service to serve up a ReactJS single page app.</p><p>One challenge is that the dashboard could end up displaying the status of hundreds of video classes. Rendering that much data would be too slow — even using a virtual DOM.</p><p>I quickly realized that each file segment should not be a div but rather a single unicode text character.</p><p>Another important optimization used on both the front and backend is the use of <a href="https://en.wikipedia.org/wiki/Bit_field">bit fields</a>. Since a class can have 5000 segments I simply create a 5000-bit field string and further reduce the required size by sending data from the backend in hex string format. The front-end converts the hex string to a binary string which is then iterated to render 0 as ◯ and a 1 as ◉.</p><p>Also once a class video is completely transferred, the segment status display is turned off to further reduce the impact of page rendering. The NPM <a href="https://www.npmjs.com/package/bitset">bitset</a> package was helpful in this regard!</p><h4>Transfer speeds</h4><p>The speed of transferring files using this approach is absolutely staggering. In one of my earlier tests, the system transferred four terabytes of data in two hours and twenty minutes — that’s <a href="https://www.wolframalpha.com/input/?i=transfer+rate+for+4TiB+in+2+hours+and+20+minutes">roughly 523MB per second</a>!</p><p>And that’s nowhere near the maximum potential. Using both a larger multi-core or cluster of multi-core machines and a higher concurrent limit of lambda invocations would yield even higher transfer speeds. Granted there are costs involved so don’t try this at home.</p><p>This post describes an approach, but one which exceeded our initial expectations. Admittedly, the task at hand is almost tailor-made for lambda — but also shows how microservices and server-less technologies can be used together. There’s no reason you can’t have both!</p><p>—</p><p><em>Thanks for reading! If you like what you read, hold the clap button below so that others may find this. You can also </em><a href="https://twitter.com/cjus"><em>follow me on Twitter</em></a><em>.</em></p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fupscri.be%2Fdde502%3Fas_embed%3Dtrue&amp;dntp=1&amp;url=https%3A%2F%2Fupscri.be%2Fhackernoon%2F&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=upscri" width="800" height="400" frameborder="0" scrolling="no"><a href="https://medium.com/media/3c851dac986ab6dbb2d1aaa91205a8eb/href">https://medium.com/media/3c851dac986ab6dbb2d1aaa91205a8eb/href</a></iframe><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a32ceef68fc5" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[A look at select Connected Fitness devices]]></title>
            <link>https://medium.com/@cjus/a-look-at-select-connected-fitness-devices-9584a683b9d3?source=rss-448274adae5e------2</link>
            <guid isPermaLink="false">https://medium.com/p/9584a683b9d3</guid>
            <category><![CDATA[iot]]></category>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[wellness]]></category>
            <category><![CDATA[fitness]]></category>
            <dc:creator><![CDATA[Carlos Justiniano]]></dc:creator>
            <pubDate>Fri, 29 May 2020 17:10:28 GMT</pubDate>
            <atom:updated>2020-05-30T03:39:24.623Z</atom:updated>
            <content:encoded><![CDATA[<p>About five years ago I joined Flywheel Sports to help build their connected Fitness Bike which like Peloton, offered riders a live and on-demand connected fitness experience featuring video and real time analytics. Some of the technical details are available on the <a href="https://medium.com/flywheel-tech">tech blog</a> and during my RedisConf 2018 and 2019 presentations.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/691/1*JpVUNhQa9sIrKFUp-Z_JGw.png" /></figure><p>In my RedisConf 2020 presentation, entitled “<a href="https://youtu.be/1SwDSm-tr-g">Creating a Model of Human Physiology w/RedisGraph</a>” I reported on how IoT and Connected Fitness devices can be utilized to transmit biometric information to cloud hosted systems where they’re ripe for data processing such as machine learning.</p><p>Connected Fitness remains a core interest of mine. During my time at Flywheel I built a number of proof-of-concept prototypes as wearable technology for connected fitness. Thus began my journey into the connected fitness space.</p><p>Many of you reading this are undoubtedly familiar with the term IoT — The Internet of Things. When IoT is applied to sports then it’s referred to as the Internet of Sports or IoS. When IoT is applied to eHealth and well being, a common term used is the Internet of Bodies or IoB. Expect to see these acronyms in widespread use in the coming years.</p><p>Throughout the years I’ve purchased a number of connected fitness devices which I’ll share in this post. The list I’ve compiled is highly opinionated and by no means comprehensive. These are just the devices I personally own and have taken a personal and professional interest in.</p><p>Lastly, it’s important to note that all of the featured products are Bluetooth connected devices which work using smart phones and in some cases, smart watches. That played a significant role in why I chose them — given my interest in building similar devices.</p><h3>Whoop band</h3><p><a href="https://www.whoop.com/">https://www.whoop.com/</a></p><p>The Whoop band is a fitness tracker which has the immediate distinction of not having an actual screen display. The tracker is used by elite athletes in the NFL and MLB. The device is worn 24/7 — and uploads bio physical data to the cloud where it’s analyzed by machine learning applications.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/347/1*Lx-th3PpPijHl-fKYl5H3Q.png" /></figure><p>One interesting use case is that it allows sport coaches to monitor their players on and off the field. For the rest of us, a key feature of the whoop band is that it tells you which days you should train harder and which days you should really be holding back — it does that by tracking your body’s recovery rate. In particular it monitors heart rate variability. That’s the amount of time between individual heart beats.</p><p>The Whoop <a href="https://www.whoop.com/thelocker/whoop-podcast-all-episodes/">podcast</a> offers a wealth of insights on the data science of fitness.</p><h3>Oura ring</h3><p><a href="https://ouraring.com/">https://ouraring.com/</a></p><p>The Oura ring is a marvel of engineering. It’s an unassuming ring you wear which contains microcontrollers and sensors. Like the Whoop, the device tracks your overall fitness readiness. Where the Oura truly excels is in its incredibly small size and its best of class sleep tracking.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/410/1*6Ag4olrdpV4stOA0WppL0Q.png" /></figure><p>A really great feature of the Oura ring is that it’s probably one of the most EMF friendly devices on the market. Using the Oura app you can stop the ring from transmitting Bluetooth data while you’re sleeping. As an aside, I’ve measure the EMF signals of various devices using my GQ EMF-390 Multi-Field EMF reader.</p><p>If you’re a developer, the Oura site offers a <a href="https://cloud.ouraring.com/docs/">programming API</a> to allow for cloud access to user fitness data.</p><h3>Smart watches</h3><p><a href="https://www.samsung.com/us/mobile/wearables">https://www.samsung.com/us/mobile/wearables</a><br><a href="https://www.fitbit.com/">https://www.fitbit.com/</a></p><p>I also use one of two smart watches, the Samsung Gear S3 and a FitBit Versa. Both the FitBit and Samsung Gear are particularly easy to program using modern web technologies such as JavaScript, HTML, CSS and SVG. What’s nice about that is that if you’re a developer you may be able to prototype your own fitness ideas.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/732/1*WOxIDqn7eU065kz5Vk-Tag.png" /></figure><h3>Muse headband</h3><p><a href="https://choosemuse.com/">https://choosemuse.com/</a></p><p>I also use a Muse headband to track my meditation practice. The device is capable of capturing Alpha, Beta, Delta, Gamma and Theta brain waves in addition to tracking head movement using accelerometer and gyroscope sensors.</p><p>An important point about the Muse is that it isn’t a toy -despite being a consumer device it’s being used in brain research at major universities such as MIT and Harvard Medical Research -and countless other research institutions around the globe.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8vdI8YohIY3wY9MCpCxsLg.png" /></figure><p>I demo’ed a technical use of the Muse headband during my RedisConf 2020 presentation. You jump directly to that <a href="https://youtu.be/1SwDSm-tr-g?t=1131">demo here</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NzzRu5guuyi7GbshbuNnSg.png" /></figure><p>I wrote a short paper entitled “<a href="https://docs.google.com/document/d/1N7II4YfoKBtdaSBbfz2aAeccy5j5mWG1_0RuU1t8oO4/edit?usp=sharing">EEG Data Acquisition using the Muse 2 headband</a>” on how this works.</p><h3>HidrateSpark</h3><p><a href="https://hidratespark.com/products/black-hidrate-spark-3">https://hidratespark.com/products/black-hidrate-spark-3</a></p><p>I keep track of how much water I’m drinking by using a HidrateSpark water bottle. This smart water bottle features a sensor that can detect the water level inside the bottle and thus measure when and how much water you’re drinking.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*T4WIbhFYdcwfKh9NSPmP0w.png" /></figure><p>The company has a free app which allows you to your hydration efforts over time and even has helpful reminders — both in app, and via the glowing effect shown above.</p><h3>Upright Posture Tracker</h3><p><a href="https://www.uprightpose.com/">https://www.uprightpose.com/</a></p><p>Then there’s my Upright posture tracking device. In our line of work where many of us are hunched over a keyboard for an ungodly number of hours good posture can help ensure proper body alignment, energy flow, and reduce future medical problems.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/466/1*h5iCqeNvPGqTslzG3x2ABw.jpeg" /></figure><p>Using the Upright I was able to determine that my posture wasn’t all that bad. My wife pointed out an important observation that when she was wearing hers, even if the device wasn’t on (in cases where it lost power) she still felt the device helped her maintain her posture — so there’s a psychological effect at play for sure.</p><p>In my own use, I definitely found that the gentle buzzing the device emits was a good reminder while using it.</p><h3>DIY Biotracking</h3><p>If you’re electronically inclined, the maker movement has a great deal of resources to help you understand how to build and work with microcontrollers to build your own DIY devices.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*YjbAyuVEU1-mZqqUE1MHUA.png" /></figure><p>This is the custom prototype I built and presented at <a href="https://youtu.be/1SwDSm-tr-g?t=1092">RedisConf 2020</a> featuring a device that captures heart rate and SPO2 data and loads it to the cloud. A nice feature of this prototype is that it’s able to connect to a WiFi router and upload data directly to the cloud via Websockets over SSL.</p><p>Along those lines I wrote about custom IoT devices <a href="https://medium.com/hackernoon/iot-for-ninjas-902c65854ab4">here</a> and <a href="https://medium.com/@cjus/my-first-printed-circuit-board-1b9fcd99a0f0">here</a>.</p><h3>Do these devices really help?</h3><p>Together these devices provide an unprecedented view at exactly how I’m maintaining and improving my health and wellness. I use most of those devices while exercising on my connected fitness equipment in my home gym.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/494/1*DWxynmLMGXdbFcLTAniidg.png" /></figure><p>While most of this sounds great — I won’t get into the concerns I have about the affects of all the bluetooth EMF I’m being exposed to.</p><p>Here are my before and after photos — taken less than six months apart. I’m not saying that these devices were largely responsible for my anti-aging efforts. What I am saying is that they can be useful in providing us with the qualitative data we can use to track our results.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/921/1*8v9SMMgr9399M2BT2yBlQQ.png" /></figure><p>In the end these products are just tools to help us focus our attention and meet our wellness goals.</p><p>In the words of the late management consultant, educator, and author, Peter Drucker: <em>“you can’t manage what you can’t measure.”</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9584a683b9d3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[My First Printed Circuit Board]]></title>
            <link>https://medium.com/@cjus/my-first-printed-circuit-board-1b9fcd99a0f0?source=rss-448274adae5e------2</link>
            <guid isPermaLink="false">https://medium.com/p/1b9fcd99a0f0</guid>
            <category><![CDATA[iob]]></category>
            <category><![CDATA[iot]]></category>
            <category><![CDATA[biometrics]]></category>
            <category><![CDATA[ehealth]]></category>
            <category><![CDATA[sensors]]></category>
            <dc:creator><![CDATA[Carlos Justiniano]]></dc:creator>
            <pubDate>Fri, 29 May 2020 14:18:13 GMT</pubDate>
            <atom:updated>2020-05-29T15:42:23.628Z</atom:updated>
            <content:encoded><![CDATA[<p>This year I had the pleasure to begin learning the fine art of printed circuit board design. Despite the hardships that the Coronavirus has caused, social isolation does have its benefits. In my case that’s translated into focused learning time.</p><p>As part of my IoT work I set out to build a connected SPo2 and HeartRate monitor which is capable of securely transmitting biometric sensor data to a cloud-hosted microservice. I demo’ed the device during my <a href="https://youtu.be/1SwDSm-tr-g?t=1243">Redis Conference presentation</a> this year on the use of IoT and Connected Fitness with RedisGraph.</p><p>To keep things relatively simple and inexpensive, I started learning a tool called <a href="https://easyeda.com/">EasyEDA</a> via Amit Rana’s <a href="https://www.udemy.com/course/pcb-design/">PCB design course on Udemy</a>. Amit did an amazing job of making the subject accessible. I completed the three hour course in just a few days.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*YM4-g9iGGJPUt8fjqZ-oLA.png" /></figure><p>Using EasyEDA, I was able to order actual boards via <a href="https://jlcpcb.com/">JLCPCB</a> right from within the software. My first three board designs suffered from my own minor design errors. But I learned more than if my first attempt would have been successful. Live and learn!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/612/1*5Zws8-6kf8XjMeZ83QZAtg.png" /></figure><p>By revision 3 I had a functioning board layout.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/925/1*dWCkRh6kl1i_TR5bg0DLyQ.png" /></figure><p>I chose to use the <a href="https://www.adafruit.com/product/3405">Adafruit HUZZAH32 — ESP32 Feather</a>, largely because of my familiarity with Adafruit components and because I wanted to use BLE to configure the microcontroller for WiFi use. I did the programming using <a href="https://platformio.org/install/ide?install=vscode">Visual Studio Code and PlatformIO</a> and along the way I took a bit of deep dive into hardware debugging and wrote <a href="https://medium.com/@cjus/advanced-microcontroller-debugging-using-the-esp-prog-a97532e38e39">Advanced Microcontroller Debugging using the ESP-PROG</a> here on Medium.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/437/1*Y4Jdl01QNOeGaVOpLyMunQ.png" /></figure><p>I used a small OLED display to include both text and graphics and a <a href="https://www.sparkfun.com/products/15219">SparkFun Pulse Oximeter and Heart Rate Sensor </a>to sample biometric data. The device sends data to a cloud hosted, containerized microservice using WebSockets over SSL.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/493/1*cCjsd1s78S_kbWW_ZYeSlA.png" /></figure><p>My first attempt at using the Sparkfun sensor was unsuccessful, perhaps due to faulty soldering. In a later attempt I used an inexpensive digital microscope to inspect solder connections.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/546/1*IZnDc0ygp6031wEvIMB6bw.png" /></figure><p>My wife designed enclosures, using our 3D printer, during the process of completing the prototype.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/957/1*pGNQbf-53tKdThqfngFK0g.png" /></figure><p>On the left is an earlier prototype which used wired connections, on the right is the smaller prototype using the completed printed circuit board.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/547/1*MBRE2djY3lJznpOpI15GGw.png" /></figure><p>I’ve written a few other articles on my IoT adventures here on Medium:</p><ul><li><a href="https://medium.com/@cjus/a-personal-iot-journey-8475dec5a7c">A personal IoT journey</a></li><li><a href="https://medium.com/hackernoon/iot-for-ninjas-902c65854ab4">IoT for Ninja’s</a></li><li><a href="https://medium.com/@cjus/hacking-alzheimers-4fc9f1d6217d">Hacking Alzheimers</a></li><li><a href="https://link.medium.com/vq9JWydQS6">Advanced Microcontroller Debugging using the ESP-PROG</a></li></ul><p>—</p><p><em>Thanks for reading! If you like what you read, hold the clap button so that others may find this. You can also </em><a href="https://twitter.com/cjus"><em>follow me on Twitter</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1b9fcd99a0f0" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Advanced Microcontroller Debugging using the ESP-PROG]]></title>
            <link>https://medium.com/@cjus/advanced-microcontroller-debugging-using-the-esp-prog-a97532e38e39?source=rss-448274adae5e------2</link>
            <guid isPermaLink="false">https://medium.com/p/a97532e38e39</guid>
            <category><![CDATA[iot]]></category>
            <category><![CDATA[development]]></category>
            <category><![CDATA[microcontrollers]]></category>
            <dc:creator><![CDATA[Carlos Justiniano]]></dc:creator>
            <pubDate>Fri, 29 May 2020 13:23:09 GMT</pubDate>
            <atom:updated>2020-05-29T15:51:05.773Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/453/1*h5a-eCW4zMen1KDNtp9C2A.png" /><figcaption>ESP-Prog board</figcaption></figure><p>Developing hardware projects using microcontrollers and sensors presents a host of challenges. For one, the fact that we use a personal computer to interface with connected devices already adds a layer of separation from the machine we actually write code in. A similar situation is present in building distributed applications over networks. Albeit without the particularities of sensors and often on devices with a relative wealth of computing resources. So that seems like a first world problem when compared to the former.</p><p>In the case of microcontrollers working with concurrent real time data we are often left with fairly inadequate methods of diagnosing behavior. I’m referring to going beyond the use of serial print / log statements and using true software and hardware debugging. The seasoned developers among you already use debuggers with the ability to set breakpoints and introspect memory and variables. Certainly, this form of debugging is considered modern when compared to the former approach. So perhaps I should have named this post Modern Microcontroller debugging. But to the community of makers coming from development on the Arduino platform the use of what we’ll cover in this post is most definitely advanced by comparison.</p><p>In this post we’ll look at debugging code on the ESP32 microcontroller using the ESP-PROG hardware debugger, both of which were created by Espressif Systems. We’ll also focus on the use of the PlatformIO IDE running on Visual Studio Code. While the information contained in this post will allow you to debug on other microcontrollers we’ll focus on the ESP32 for two reasons:</p><ul><li>We need an actual MCU for practical discussion.</li><li>The ESP32 is the MCU I’m using in my own work. The actual board we’ll use is the Adafruit Hazzah32 which features the Espressif ESP32-WROOM-32 MCU.</li></ul><p>All too often articles like this are written where the reader is unclear what motive the author had in sharing the post. In an earlier life, my work featured the use of X86 assembler language and hardware debuggers such as the early <a href="https://en.wikipedia.org/wiki/SoftICE">SoftIce</a> in the early 90s. Along the way I’ve used C/C++ debuggers from GDB to the Microsoft C IDE version 1.x and beyond. So basically low-level debugging is what I’ve come to expect from the programming tools I’ve used. My current work focuses on IoT devices in the emerging Internet of Bodies (IoB) and Internet of Sports (IoS) platforms in the Health, Fitness and Wellness space. Also, the IoT devices I build communicate with cloud-hosted microservices. So given the overall end-to-end complexity — there’s serious incentive to go as low as one needs to while debugging. I wrote this post to share my own experience in getting this to work.</p><p>Ok, with all that out of the way let’s dive right in.</p><h3>The ESP-Prog board</h3><p>The ESP-Prog is a development and debugging tool which connects to your machine (PC/Laptop) and an ESP32 board to fully control it. By control, I mean that you step through program code using your laptop (running PlatformIO) and use it to send commands to the ESP-Prog board which in turn controls the execution of code running on the ESP32. This all happens over USB to the ESP-Prog and over a few wires from the ESP-Prog to the ESP32 board. We’ll take a closer look at this wiring in just a bit.</p><p>I was working with the ESP8266 and ESP32 for a few weeks before I encountered Andreas Spiess’s YouTube channel and video <a href="https://www.youtube.com/watch?v=psMqilqlrRQ">introducing Inline Debugging for ESP32</a> Andreas discussed two hardware devices for inline debugging, the Olimex ARM-USB-OCD and the Espressif ESP-Prog board. The ESP-Prog was both cheaper and specifically designed by Espressif — so it was the obvious choice. I purchased a board from <a href="https://www.digikey.com/product-detail/en/ESP-PROG/1965-ESP-PROG-ND/10259352">Digi-Key</a> for just $13 bucks.</p><p>Upon closer examination we can see that there are a number of interesting components on the ESP-Prog board. On the left we see that there’s a Micro USB connection where you’ll connect to your PC or Laptop. There are also buttons for manually booting and resetting the device. Just to the right of the USB connector is a USB Bridge chip. That’s the chip that converts USB to serial data. This is necessary because the interface from the ESP-Prog to the ESP32 board is over serial communication using the UART interface for transmit and receive.</p><p>On the lower right we see the JTAG 2.54 mm connector. We’ll use those pins to wire up the ESP32.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/860/1*4qCM3bQX0dMnYOGkiofUUw.png" /></figure><p>You might be wondering what this JTAG thing is anyway? JTAG is an industry standard for working with circuit boards. It’s essentially a method of interfacing hardware for the purpose of testing. JTAG itself stands for the Joint Test Action Group which created the standard that was later codified by the IEEE in 1990 in a standard entitled “Standard Test Access Port and Boundary-Scan Architecture”. You can find out more about JTAG here <a href="https://en.wikipedia.org/wiki/JTAG">https://en.wikipedia.org/wiki/JTAG</a></p><p>If we rotate the JTAG 2.54 mm connector clockwise we can see the following pin arrangement.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/370/1*1wSHpd-OjlKajoOn5Grq3A.png" /></figure><p>We’ll only use six of the ten pins: 1, 2, 3, 4, 6, 8</p><h3><strong>Wiring</strong></h3><p>As is common for me in these sort of projects, I tend to create visual aids to ensure I get the wiring right. Despite this extra care, I’d be lying if I told you that doing this has prevented me from having fried a board or two.</p><p>This table maps pins on the Adafruit Huzzah32 to the ESP-PROG using colored wires.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/414/1*-D0EBE4MWjqWD25RVofO2w.png" /></figure><p>On the bottom of the Huzzah32 we see the pin labelling which I’ve colored to match the wiring we’ll use.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/546/1*lttomYeOgSUyq2nsiZiuOg.png" /></figure><p>Here is the actual wiring on a broad board. I added an OLED display as an extra debugging tool. We’ll step through the OLED display driver code later in this post.</p><p>This Fritzing diagram below should clarify the wiring. The female header on the left represents the JTAG 2.54 mm connection we discussed earlier. The labels on the Huzzah32 might seem a bit misleading but they’re correct.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7uepvmC0Ywz7zJ0iZQA--Q.png" /></figure><h4><strong>Now for the software setup</strong></h4><p>Once we have the wiring complete we’ll need to ensure we have the correct USB drivers installed on our development machine.</p><p><strong>USB drivers</strong></p><p>The ESP-Prog features a USB Bridge Controller chip provided by FTDI. So we’ll need to install <a href="https://www.ftdichip.com/Drivers/VCP.htm">FT2232HL (VCP) Virtual COM Port driver.</a></p><p><strong>TODO </strong>Summarize this section: <a href="https://github.com/espressif/esp-iot-solution/blob/master/documents/evaluation_boards/ESP-Prog_guide_en.md#321-the-working-mode-of-usb-bridge">https://github.com/espressif/esp-iot-solution/blob/master/documents/evaluation_boards/ESP-Prog_guide_en.md#321-the-working-mode-of-usb-bridge</a></p><p>I’m using a Mac so I had to install the Mac OS X 10.9 and above version of the driver.</p><p>* USB drivers for the ESP-Proj</p><p>* problems on Mac</p><pre>sudo kextunload -b com.FTDI.driver.FTDIUSBSerialDriver</pre><p><strong>PlatformIO and Visual Studio Code</strong></p><p>I used the Arduino IDE for years despite the degree to which it’s lacking in as a modern IDE. My dislike of the Eclipse IDE kept it from ever being a viable alternative. Fortunately, I learned a while back that I could use my editor of choice, Visual Studio Code with an add-on called PlatformIO. The resulting combination forms a world class tool for working with microcontrollers! And as my best friend would say: “and the price is right — Free!”.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/711/1*Df8tJy64oPckrTNtbRQTxg.png" /></figure><p>There are lots of getting started with Visual Studio Code tutorials so I won’t spend time on that in this article. However, you can download <a href="https://code.visualstudio.com/">Visual Studio Code</a> for free and proceed to watch some <a href="https://www.youtube.com/results?search_query=visual+studio+code">YouTube videos</a>.</p><p>Once you have Visual Studio Code installed you can easily add the free PlatformIO add-on using the instructions provided on the <a href="https://platformio.org/">PlatformIO</a> website. Here again, you’ll find lots of great <a href="https://www.youtube.com/results?search_query=platformio">PlatformIO tutorials on YouTube</a>.</p><h3>Using PlatformIO</h3><p>After installing Visual Studio Code and PlatformIO we’ll need to configure the libraries we’ll need in order to work with ESP32 boards.</p><p><strong>TODO</strong> Add configuration screens</p><h3>Debugging</h3><p>With the ESP-Prog wired and connected to an ESP32 board, we’re able to load a sample program and try advanced debugging.</p><pre>serial.serialutil.SerialException: [Errno 16] could not open port /dev/cu.usbserial-14100: [Errno 16] Resource busy: &#39;/dev/cu.usbserial-14100&#39;</pre><p><strong>TODO</strong> Show stepping through code screenshots</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/841/1*PFpmWETJKmTL-CYyO6pzYA.png" /></figure><p><strong>TODO</strong> Show a cleaner version of the project using custom cables</p><h3><strong>Resources</strong></h3><ul><li><a href="https://github.com/espressif/esp-iot-solution/blob/master/documents/evaluation_boards/ESP-Prog_guide_en.md">Introduction to the ESP-Prog Board</a></li><li><a href="https://www.ftdichip.com/Support/Documents/AppNotes/AN_134_FTDI_Drivers_Installation_Guide_for_MAC_OSX.pdf">FTDI Drivers Installation guide for MAC OS X</a></li><li><a href="https://www.ftdichip.com/Drivers/VCP.htm">FT2232HL (VCP) Virtual COM Port driver</a></li><li><a href="https://www.youtube.com/watch?v=Ir2PVz1870E">Installing D2xx drivers on macOS</a></li><li><a href="https://code.visualstudio.com/">Visual Studio Code</a></li><li><a href="https://platformio.org/">PlatformIO</a></li><li><a href="https://docs.platformio.org/en/latest/plus/debug-tools/esp-prog.html">PlatformIO PIO Unified Debugger ESP-Prog documentation</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a97532e38e39" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[A select list of my open sourced projects]]></title>
            <link>https://medium.com/@cjus/a-select-list-of-my-open-sourced-projects-fecac85ba9ca?source=rss-448274adae5e------2</link>
            <guid isPermaLink="false">https://medium.com/p/fecac85ba9ca</guid>
            <category><![CDATA[github]]></category>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[hacking]]></category>
            <category><![CDATA[open-source]]></category>
            <dc:creator><![CDATA[Carlos Justiniano]]></dc:creator>
            <pubDate>Sun, 28 Jul 2019 17:44:10 GMT</pubDate>
            <atom:updated>2019-07-28T17:44:10.191Z</atom:updated>
            <content:encoded><![CDATA[<p>This afternoon I decided to take a stroll down memory lane and comb through my Github repositories.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*LCjGHN-1tyzdNPjpYuwgqA.jpeg" /><figcaption>Credit: Getty Images</figcaption></figure><p>I found projects dating back to 2010. I’ve certainly been programming long before then… but didn’t join Github until 2010. The projects below use languages such as C/C++, C#, Python and JavaScript. Projects range from microcontroller based Arduino projects to NodeJS-based Microservice libraries. Some of the projects are even extensively documented. Imagine that? Dude must have a lot of time on this hands, right? There’s almost something there for everyone.</p><h4>UDP server to receive OSC streaming messages from the Interaxon Muse headband</h4><p>Jul 14, 2019<br><a href="https://github.com/cjus/brain-wave-tracker">https://github.com/cjus/brain-wave-tracker</a></p><h4>Hacking Alzheimers using Light Therapy</h4><p>Jun 5, 2019<br><a href="https://github.com/cjus/brainlights">https://github.com/cjus/brainlights</a></p><h4>Pack and unpack JSON files</h4><p>Jan 25 2019<br><a href="https://github.com/cjus/simple-json-pack">https://github.com/cjus/simple-json-pack</a></p><h4>Web-based Time Hacker Clock</h4><p>Oct 1, 2018<br><a href="https://github.com/cjus/web-thc">https://github.com/cjus/web-thc</a></p><h4>Arduino-based Time Hacker Clock</h4><p>Sep 29, 2018<br><a href="https://github.com/cjus/arduino-thc">https://github.com/cjus/arduino-thc</a></p><h4>RaspberryPi-based — Time Hacker Clock</h4><p>Sep 21, 2018<br><a href="https://github.com/cjus/raspi-thc">https://github.com/cjus/raspi-thc</a></p><h4>Network Request Mob — load generation tool</h4><p>Mar 15, 2018<br><a href="https://github.com/cjus/requestmob">https://github.com/cjus/requestmob</a></p><h4>JMdb — A JSON Memory Database with an RESTFul API</h4><p>May 14, 2017<br><a href="https://github.com/cjus/jmdb">https://github.com/cjus/jmdb</a></p><h4>Neo4j adapter with Redis caching</h4><p>Dec 8, 2016<br><a href="https://github.com/cjus/neo4j-redis">https://github.com/cjus/neo4j-redis</a></p><h4>Hydra — A light-weight library for building distributed applications such as microservices</h4><p>Oct 4, 2016<br><a href="https://github.com/flywheelsports/hydra">https://github.com/flywheelsports/hydra</a></p><h4>HydraExpress — A module which wraps Hydra and ExpressJS into a library for building distributed applications — such as microservices</h4><p>Oct 4, 2016<br><a href="https://github.com/flywheelsports/hydra-express">https://github.com/flywheelsports/hydra-express</a></p><h4>HydraRouter — A service aware router for Hydra Services. Implements an API Gateway and can route web socket messages</h4><p>Oct 4, 2016<br><a href="https://github.com/flywheelsports/hydra-router">https://github.com/flywheelsports/hydra-router</a></p><h4>Hydra-Cli — a command line tool for Hydra</h4><p>Aug 5, 2016<br><a href="https://github.com/flywheelsports/hydra-cli">https://github.com/flywheelsports/hydra-cli</a></p><h4>JSON Tree is a module for manipulating and querying JavaScript objects</h4><p>Apr 5, 2016<br><a href="https://github.com/cjus/json-tree">https://github.com/cjus/json-tree</a></p><h4>Universal Messaging Format</h4><p>Mar 29, 2016<br><a href="https://github.com/cjus/umf">https://github.com/cjus/umf</a></p><h4>Bunyan Extractor — extract data from bunyan log files</h4><p>Oct 26, 2014<br><a href="https://github.com/cjus/bunext">https://github.com/cjus/bunext</a></p><h4>Modern web application seed project</h4><p>Sep 12, 2014<br><a href="https://github.com/cjus/modern-web-seed">https://github.com/cjus/modern-web-seed</a></p><h4>A q/promise based pure JS library for working with Neo4j</h4><p>Aug 20, 2014<br><a href="https://github.com/cjus/qcypher">https://github.com/cjus/qcypher</a></p><h4>WebSocket Universal Message Router</h4><p>Nov 2, 2013<br><a href="https://github.com/cjus/umf_router">https://github.com/cjus/umf_router</a></p><h4>msgCourier Whirl Wind — Load Testing and Message Benchmarking Tool</h4><p>Sep 4, 2010<br><a href="https://github.com/cjus/mcww">https://github.com/cjus/mcww</a></p><h4>Section34 Multimedia Engine (Sprite Animation Library for Win32)</h4><p>Sep 4, 2010<br><a href="https://github.com/cjus/s34mme">https://github.com/cjus/s34mme</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fecac85ba9ca" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Top 10 posts I’ve authored on Medium]]></title>
            <link>https://medium.com/@cjus/top-10-posts-ive-authored-on-medium-734e464bf60f?source=rss-448274adae5e------2</link>
            <guid isPermaLink="false">https://medium.com/p/734e464bf60f</guid>
            <category><![CDATA[blogging]]></category>
            <category><![CDATA[technology]]></category>
            <dc:creator><![CDATA[Carlos Justiniano]]></dc:creator>
            <pubDate>Sun, 28 Jul 2019 16:43:09 GMT</pubDate>
            <atom:updated>2019-07-31T17:00:36.176Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yU10x1LAIo4pAn9sTfTE8g.jpeg" /><figcaption>credits: <a href="https://www.istockphoto.com/portfolio/Christian_Horz?mediatype=photography">Christian Horz</a> via iStockPhoto</figcaption></figure><p>The following posts are the top 10 most popular articles I’ve written on Medium. The topics range from programmer productivity tips to how-tos ranging from the use of microcontollers to cloud-based infrastructure.</p><h4>#1 at 21K views: <strong>The Time Hacker Method</strong></h4><p>A time management and productivity article.</p><p><a href="https://medium.com/hackernoon/the-time-hacker-method-12970c47f04f">The Time Hacker Method</a></p><h4>#2 at 20K views: <strong>Handy Docker Aliases</strong></h4><p>Workflow productivity tips for working with Docker.</p><p><a href="https://medium.com/hackernoon/handy-docker-aliases-4bd85089a3b8">Handy Docker Aliases</a></p><h4>#3 at 13.1K views: <strong>Installing Docker CE on an AWS EC2 instance running Ubuntu 16.04</strong></h4><p><a href="https://medium.com/@cjus/installing-docker-ce-on-an-aws-ec2-instance-running-ubuntu-16-04-f42fe7e80869">Installing Docker CE on an AWS EC2 instance running Ubuntu 16.04</a></p><h4>#4 at 10.1K views: <strong>Building Light-weight Microservices using Redis</strong></h4><p><a href="https://medium.com/flywheel-tech/building-light-weight-microservices-using-redis-23f051624647">Building Light-weight Microservices using Redis</a></p><h4>#5 at 9.94K views: <strong>Desktop apps using Electron, Preact and Material Design</strong></h4><p><a href="https://medium.com/@cjus/desktop-apps-using-electron-preact-and-material-design-8161938624c6">Desktop apps using Electron, Preact and Material Design</a></p><h4>#6 at 7.6K views: <strong>Raspberry Pi based Wifi Routers</strong></h4><p>How-to instructions for building your very own Wifi Router using a $5 Raspberry Pi.</p><p><a href="https://medium.com/hackernoon/raspberry-pi-based-wifi-routers-3b8c7c49143f">Raspberry Pi based Wifi Routers</a></p><h4>#7 at 6.6K views IoT for Ninjas</h4><p>Using microcontrollers to power Connected Fitness</p><p><a href="https://medium.com/hackernoon/iot-for-ninjas-902c65854ab4">IoT for Ninjas</a></p><h4>#8 at 4.6K views: <strong>Hydra, AWS, and Docker</strong></h4><p>A NodeJS based microservice deployment story.</p><p><a href="https://medium.com/hydramicroservices/hydra-aws-and-docker-a-microservices-deployment-story-ce7c3f4abad9">Hydra, AWS, and Docker: a microservices deployment story</a></p><h4>#9 at 4.2K views: Time Hacker Clocks</h4><p><a href="https://medium.com/@cjus/time-hacker-clocks-3a1491dd02a7">Time Hacker Clocks</a></p><h4>#10 at 1.3K views: Cloud-based file transfer at scale</h4><p><a href="https://medium.com/hackernoon/cloud-based-file-transfer-at-scale-63d8e2dacb3a">Cloud-based file transfer at scale</a></p><p>A personal favorite (and one dear to my heart) is a relatively recent post on building light emitting devices for use in experimental Alzheimers treatments. The article includes plans for a working prototype device based on an MIT research study.</p><p><a href="https://medium.com/@cjus/hacking-alzheimers-4fc9f1d6217d">Hacking Alzheimers</a></p><p>The “views” stats shown above were current as of July 2019.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=734e464bf60f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Hacking Alzheimers]]></title>
            <link>https://medium.com/@cjus/hacking-alzheimers-4fc9f1d6217d?source=rss-448274adae5e------2</link>
            <guid isPermaLink="false">https://medium.com/p/4fc9f1d6217d</guid>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[alzheimers]]></category>
            <category><![CDATA[brain]]></category>
            <category><![CDATA[biohacking]]></category>
            <category><![CDATA[arduino]]></category>
            <dc:creator><![CDATA[Carlos Justiniano]]></dc:creator>
            <pubDate>Fri, 07 Jun 2019 15:56:21 GMT</pubDate>
            <atom:updated>2019-06-07T16:37:00.216Z</atom:updated>
            <content:encoded><![CDATA[<p>Building DIY light therapy devices</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*UVexxXe47Kv5KYfs7NGz2w.jpeg" /><figcaption>Image by Gerd Altmann via Pixabay</figcaption></figure><h3>Alzheimers hitting home</h3><p>The widespread impact of Alzheimers virtually guarantees that you or someone you know has a loved one who has suffered from Alzheimers. For me it’s been several people I’m very close to, my grandmother, mother, mother-in-law and my father-in-law’s life partner. While that seems a bit much to me, I’m far from alone.</p><p>In a video published in December 2016 on YouTube, the <br>Massachusetts Institute of Technology (MIT) shared some alarming statistics:</p><ul><li>One in three adults in the United States will die from Alzheimers or age related dementia.</li><li>There are now 46 million people with Alzheimers worldwide.</li><li>The total societal costs associated with age related dementia will reach 2 trillion by 2030.</li><li>Alzheimers expenses are the greatest direct healthcare cost to the United States economy — greater than cancer and heart disease.</li></ul><h3>Promising Research</h3><p>I year or so ago, I came across Torin Blankensmith’s project: <a href="https://github.com/torinmb/Alzheimers_Flashing_Gamma_Lights">Alzheimers Flashing Gamma Lights</a> which he based on a WNYC Studios broadcast entitled <a href="https://www.wnycstudios.org/story/bringing-gamma-back">Bringing Gamma Back</a>. The broadcast featured an <a href="http://news.mit.edu/2016/visual-stimulation-treatment-alzheimer-1207">MIT study</a> which showed a marked improvement in laboratory mice suffering from early stage Alzheimer when they were given an hour long session of pulsating light at 40hz.</p><p>I’ve wanted to try implementing Torin’s work with Arduino Microcontrollers ever since.</p><p>You might be wondering what light has to do with Alzheimers?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*FhTiyiHKg2OW0Xh6Twnfqg.png" /><figcaption><a href="https://www.youtube.com/watch?v=O_p4QWkE2Ls">Light-based therapy for Alzheimer’s disease</a></figcaption></figure><p>As you probably already know, your brain has billions of cells called Neurons which function using a <a href="https://en.wikipedia.org/wiki/Bioelectricity">bio-electrical</a> process.</p><p>It’s believed that Alzheimers disease disrupts brain signalling. As neurons fire they create a specific rhythm which has been measured at a frequency of 40 Hz, a frequency range known as the gamma frequency.</p><p>Researchers have discovered that gamma rhythm is necessary to activate the brain’s cleanup genes called <a href="https://en.wikipedia.org/wiki/Microglia">Microglia</a>. Microglia function as part of the brains immune system.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/499/1*9uf7WphHuLWSoFUfxOnccQ.jpeg" /><figcaption>Activated Microglia via Wikipedia</figcaption></figure><p>When gamma rhythm is reduced the microglia are less active and thus less engaged in cleanup. The absence of cleanup, leads to the build up of a protein called Amyloid Beta which results in less gamma rhythms. The end result is reduced brain function such as the ability to recall memories.</p><p>For a clearer description see this five minute <a href="https://www.youtube.com/watch?v=O_p4QWkE2Ls">YouTube</a> video.</p><h3>Commercial product</h3><p>I managed to find one company offering <a href="https://gammalighttherapy.com/">commercial products</a>. I’m sure there are more. That said, this post is for the biohacker / DYIer.</p><h3>Design</h3><p>Since Torin’s project is several years old, I’ve upgraded the components and overall design. For example, the original product didn’t specify a goggle and instead used room lights. I felt more direct stimulation might be more effective. I also changed the microcontroller and wrote a new program which drives more modern devices. Lastly I assembled build instructions for a sharp looking product.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/656/1*LYJdTgN6RAlx93xLXx1o6Q.gif" /></figure><p>My version is based on the Adafruit Trinket M0 an extremely small board featuring the Atmel ATSAMD21E18 microcontroller. The machine is programmable using the Arduino IDE and Adafruit board definition files.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/489/1*zXz5mu7vFLNTBn_5gUGJFA.png" /><figcaption>Here’s the Adafruit Trinket M0 next to a dime for reference</figcaption></figure><p>The overall design is fairly straightforward featuring a microcontroller board, textile button switch and two micro RGB LEDs in series.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jwdYhi6wKOHy7tECT6cKmw.png" /></figure><p>I used heat sink tubing to wrap wired connections and form an enclosure for both the LED pair as well as the microcontroller board.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1013/1*IQ3wD1WAYAZbZHo35E7P1A.png" /></figure><p>Here’s a closeup of the microcontroller being wrapped in a heat sink tube. The nice thing about the heat sink wrap is that it secures fragile soldered connections while also making the component easier to glue mount.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xg3ryLPwQ7nPViBs4FRXDg.png" /></figure><p>I was able to mount the board on the left side of Dewalt safety googles.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yIS4zV-7XYYtXAA8VN7wZw.jpeg" /></figure><p>I then crazy glued the light strip onto the inside of the googles. A micro-usb cable provides the board with power. The Trinket M0 board has support for mounting a JST-PH 2-Pin SMT Right Angle Connector for adding a lithium battery.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IQ_8qSar4_kAEt9Dnxjrjw.jpeg" /></figure><h3>Code</h3><p>The programming for the microcontroller is surprisingly simple. As a small embellishment I use the textile button presses to trigger LED color changes from white, to red to green and then blue.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/380943a58544b8801e12f3412b870dc6/href">https://medium.com/media/380943a58544b8801e12f3412b870dc6/href</a></iframe><h3>Additional info on YouTube</h3><p>If you’d like to learn more about the science behind light based therapy see this YouTube video and related videos.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FO_p4QWkE2Ls%3Ffeature%3Doembed&amp;url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DO_p4QWkE2Ls&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FO_p4QWkE2Ls%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/afb90420d9df3222f53d8b04b1eeab5e/href">https://medium.com/media/afb90420d9df3222f53d8b04b1eeab5e/href</a></iframe><h3>Parts</h3><p>Review the links below if you’re interested in building your own prototype.</p><ul><li><a href="https://www.adafruit.com/product/3500">Adafruit Trinket M0</a></li><li><a href="https://www.adafruit.com/product/367">Tactile Button switch (6mm) x 20 pack</a></li><li><a href="https://www.amazon.com/dp/B01D1FFVOA/ref=cm_sw_r_cp_ep_dp_5E4TBbGTXMX53">ALITOVE 100pcs WS2812B Addressable 5050 smart RGB LED Pixel light on Black Heat Sink PCB Board</a></li><li><a href="https://www.amazon.com/dp/B000RKQ1O2/ref=cm_sw_r_tw_dp_U_x_mVy-CbXJHTME5">Dewalt DPG82–21 Concealer Smoke Anti-Fog Dual Mold Safety Goggle</a></li><li><a href="https://www.amazon.com/dp/B000AL6WLA/ref=cm_sw_r_tw_dp_U_x_tXR6CbPF1AEZY">Permatex 81158 Black Silicone Adhesive Sealant, 3 oz. Tube</a></li><li>Heat shrink tubing</li></ul><p>Optional:</p><ul><li>JST-PH 2-Pin SMT Right Angle Connector <a href="https://www.adafruit.com/product/1769">https://www.adafruit.com/product/1769</a></li><li>Tiny Lithium Ion Polymer Battery — 3.7v 100mAh <a href="https://www.adafruit.com/product/1570">https://www.adafruit.com/product/1570</a></li></ul><p><em>Thanks for reading! If you like what you read, hold the clap button below so that others may find this. You can also </em><a href="https://twitter.com/cjus"><em>follow me on Twitter</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4fc9f1d6217d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Vanity license plates — a geek perspective]]></title>
            <link>https://medium.com/@cjus/vanity-license-plates-a-geek-perspective-12e0bb482ad6?source=rss-448274adae5e------2</link>
            <guid isPermaLink="false">https://medium.com/p/12e0bb482ad6</guid>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[development]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[algorithms]]></category>
            <category><![CDATA[naturallanguageprocessing]]></category>
            <dc:creator><![CDATA[Carlos Justiniano]]></dc:creator>
            <pubDate>Tue, 21 May 2019 21:18:23 GMT</pubDate>
            <atom:updated>2019-05-21T21:26:15.256Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OxRyOGlTQ_M8wE3PyUUJbw.jpeg" /><figcaption>Photo by <a href="https://unsplash.com/photos/JK45WYC7c0Y?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Jesse Borovnica</a> on <a href="https://unsplash.com/search/photos/license-plate?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Unsplash</a></figcaption></figure><h3>Vanity license plates — a geek perspective</h3><p><em>This is a post I wrote back in 2010. The problem remains interesting almost a decade later.</em></p><p>Recently while driving to work I noticed several vanity license plates. I found myself contemplating how a computer could be used to decipher the meaning behind those plates. Plates reading NICEBODY and ATOYOTA seem easy to decipher because they’re just words without spaces. However, HYPRDRV (hyper drive) and TIH2 HO (mirror image of “oh shit”) are more difficult. Even more complicated would be plates like PD4BYEX (Paid for by Ex), W84ME (Wait for me).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/280/1*tfoGjtcM0fNQREgBgNTbtA.gif" /></figure><p>From the examples we just considered we can see that there are several types of challenges. Even the case that appears to be the simplest, words without spaces, is actually non-trivial. How do you know where one word begins and ends? One method might be to simply search for sub matches. For example, ATOYOTA would result in a match for the first character A, then TOY and later TOYOTA. NICEBODY, has NICE, ICE and BODY.</p><p>Some plates such as PD4BYEX (paid for by ex) require searching for suitable character to word substitutions. We’d have to consider vowel removal: PAID resulting in PD and prefixes such as EX (ex-spouse).</p><p>In an advanced solution words could be compared semantically to rank potential matches. Consider W84ME might result in [Wait four me], [W8 four me], [Wine 84 me]. Scoring would be higher for legitimate phrases. Legitimate phrases would be those which fared well against a Natural Language Parts of Speech tagger.</p><p>The output of such a software program would still be a list of ranked potential matches. Meaning that the user would have to weed through results to select what they believe to be the most likely match.</p><p>Another approach might be a search engine with a crowd sourcing option that allows users to contribute and select (possibly rank) potential solutions. The results of user interactions could be used to statistically train the search engine.</p><p>If packaged into a mobile app this would be fun for those long family road trips. At a social networking extreme users could get social points for having spotted the most interesting vanity plates. I can just see the twitter post now: @Ryan just spotted IMABRAT (I am a brat) on I280. This would undoubtably require capturing license plate photos and having them uploaded to a cloud based server solution for character recognition, search and other backend processing.</p><p>Interesting problem. I’m sure we’ll see similar apps soon.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=12e0bb482ad6" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What’s your story?]]></title>
            <link>https://medium.com/@cjus/whats-your-story-6ef75ed1f4ec?source=rss-448274adae5e------2</link>
            <guid isPermaLink="false">https://medium.com/p/6ef75ed1f4ec</guid>
            <category><![CDATA[tech]]></category>
            <category><![CDATA[career-advice]]></category>
            <category><![CDATA[hiring]]></category>
            <category><![CDATA[technology]]></category>
            <dc:creator><![CDATA[Carlos Justiniano]]></dc:creator>
            <pubDate>Tue, 21 May 2019 16:10:05 GMT</pubDate>
            <atom:updated>2019-05-21T16:10:05.417Z</atom:updated>
            <content:encoded><![CDATA[<h4>Career advice for developers</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*JMsN-rnyB1H2q14RbQsVBw.jpeg" /><figcaption>Photo by <a href="https://unsplash.com/photos/4m7gmLNr3M0?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">David Rangel</a> on <a href="https://unsplash.com/search/photos/code?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Unsplash</a></figcaption></figure><p><em>I posted this story back in 2010 on my personal blog. The advice is still valid today.</em></p><p>We’re currently in the process of interviewing candidates for three open positions in our software development group. This process typically takes us several months on average. We see several dozen people during the process of weeding through a large pool of candidates. Our goal is simply to hire the best candidate for the job — and perhaps we take that too seriously.</p><p>Two questions I always pose to candidates are “What’s your story?” and “What have you done?”. I’m always interested in hearing about the paths that people take during their careers. Sadly, most people are unprepared for such direct inquires. Nine times out of ten people refer to their most recent experiences. Are their careers so boring that they can’t remember what they did five years ago? I don’t think so. At least not in all cases. The issue may be that few people think of their careers as an unfolding story. They’re unable to weave together the events that lead them to this point in time.</p><p>When asked, “Have you worked with such-n-such technology?”, candidates often answer: “Oh, I didn’t have an opportunity to use that in my last job”. What? Who waits for opportunities to appear at their door steps? Apparently many people do. The successful ones are those who spend some of their spare time exploring new technologies and ideas. Consequently, they’re typically the types of people who have interesting stories to tell. One of my favorite and well worn t-shirts reads: “Carpe Noctem — Seize the Night”.</p><p>One employer I once met had a saying, “We try to hire people with the right DNA”. I’ve come to understand what he meant and perhaps I’ve been guilty of doing the same. It’s been my experience that software development is a tough business. Many people aren’t suited for work in the field — even many of the seemingly talented developers we’ve all met.</p><p>Sooner or later you’ll meet someone who is interested in more than what’s on your resume. Perhaps this has already happened?</p><p>Here’s my advice:</p><blockquote><em>* If you don’t have a genuine interest or some degree of passion for what you do then you may be wasting your most valuable asset. Consider a career that will make you happy.<br>* Reflect on the experiences you’ve had in your career. How have the twists and turns helped to refine your views and interests? What sorts of patterns are visible?<br>* Document your story. A blog is a great way to do this. It’s what I’ve tried to do with this site.<br>* Carpe Noctem — Seize the Night. It’s never too late to refine your story.</em></blockquote><p>Doing any of this will separate you from the pack and seriously increase your earning potential.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6ef75ed1f4ec" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>