<?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 Tina Tran on Medium]]></title>
        <description><![CDATA[Stories by Tina Tran on Medium]]></description>
        <link>https://medium.com/@tinatran._?source=rss-29f1333fd966------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*S9zz4XSwjUA6X269JFdH3Q.jpeg</url>
            <title>Stories by Tina Tran on Medium</title>
            <link>https://medium.com/@tinatran._?source=rss-29f1333fd966------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 19:38:21 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@tinatran._/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[Building an Ethereum Wallet in Node.js]]></title>
            <link>https://medium.com/@tinatran._/building-an-ethereum-wallet-in-node-js-8b96ce51b748?source=rss-29f1333fd966------2</link>
            <guid isPermaLink="false">https://medium.com/p/8b96ce51b748</guid>
            <category><![CDATA[javascript]]></category>
            <category><![CDATA[crypto-wallet]]></category>
            <category><![CDATA[web3]]></category>
            <category><![CDATA[ethereum-blockchain]]></category>
            <category><![CDATA[blockchain]]></category>
            <dc:creator><![CDATA[Tina Tran]]></dc:creator>
            <pubDate>Thu, 12 Mar 2026 03:38:20 GMT</pubDate>
            <atom:updated>2026-03-12T03:38:20.860Z</atom:updated>
            <content:encoded><![CDATA[<p>Creating a simple Ethereum wallet that generates addresses and reads balances from the blockchain.</p><h3>Getting Started (My First Roadblock)</h3><p>At the start of this project, my goal was pretty simple: I wanted to build a small prototype so I could actually understand how crypto wallets work. Instead of just reading docs or watching tutorials, I wanted to build something myself and see how everything fits together.</p><p>My original plan was to build the project using Replit. It seemed like the easiest way to get started since you can write and run code directly in the browser without setting up anything locally.</p><p>But things didn’t go as smoothly as I expected.</p><p>When I opened the project in Replit, my GPT Focus Coach that my TKS director programmed kept telling me to find the code. I coulnd’t find it so I switched to doing the project on the There were a bunch of panels, menus, and AI tools, but I couldn’t clearly find where the files were or where I was supposed to edit the JavaScript code.</p><p>At first I thought I was just missing something obvious. I clicked through different menus trying to find the editor or project files, but I kept ending up in different UI sections that didn’t look like the normal project structure I expected.</p><p>It got really frustrating pretty quickly and I really wanted to quit.</p><p>The goal of the project was to learn how wallets work and interact with the blockchain, not spend a ton of time figuring out a platform interface.</p><p>So eventually I decided it would be easier to just switch approaches and run the project on my own computer.</p><p>This turned out to be the right move.</p><p>Once I created a local Node.js project, everything became much clearer. I could see the folder structure, edit the files directly, and run the program from the command line.</p><p>Working locally also helped me understand what was actually happening behind the scenes. I could run the server, edit the index.js file, and instantly see the changes when I refreshed the page.</p><p>Even though that detour slowed the start of the project a little, it ended up being a good learning moment. It reminded me that sometimes switching tools is the fastest way forward when something isn’t working.</p><p>Once the project was set up locally, I could finally focus on the actual goal: building a working Ethereum wallet prototype.</p><h3>Understanding How Ethereum Wallets Work</h3><p>Before building the wallet itself, I wanted to understand what an Ethereum wallet actually is.</p><p>One big misconception is that crypto wallets “store” cryptocurrency. They actually don’t.</p><p>All balances and transactions exist on the blockchain itself. A wallet just manages the keys that allow you to control funds tied to a specific blockchain address.</p><p>There are two main parts to this:</p><p><strong>Public Address</strong></p><p>This is the wallet address that people can send crypto to. It’s similar to a bank account number. Ethereum addresses usually start with 0x.</p><p><strong>Private Key</strong></p><p>This is the secret key that proves you own the wallet. Whoever has the private key can control the funds associated with that address.</p><p>Because of this, private keys need to stay secure. If someone gets access to your private key, they basically control your wallet.</p><p>Most wallet apps hide all this complexity behind a simple interface. But under the hood they’re doing a lot of cryptographic work to generate keys and sign transactions.</p><p>For this project, I used ethers.js, which handles most of these cryptographic operations.</p><p>It makes things like generating wallets, connecting to the blockchain, and reading balances much easier.</p><h3>Setting Up the Project</h3><p>After deciding to run the project locally, the next step was setting up a basic Node.js environment.</p><p>Node.js is useful here because it lets you run JavaScript outside the browser and build backend applications. Since the wallet needed to interact with the Ethereum network, running a small server made things easier.</p><p>First I initialized a new Node project:</p><pre>npm init -y</pre><p>This creates a package.json file that manages the project dependencies.</p><p>Next I installed the library that would handle the blockchain interaction:</p><pre>npm install ethers</pre><p>After that I created the main project file called index.js.</p><p>This file contains the server code and the wallet logic.</p><p>Inside it, the project imports the modules it needs:</p><pre>const { ethers } = require(&quot;ethers&quot;);<br>const http = require(&quot;http&quot;</pre><p>At this point the basic structure of the project was ready, and I could start implementing the actual wallet functionality.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8cDAxQdLqHReb4gGaQwHWw.png" /><figcaption>Local Node.js project structure used for the Ethereum wallet prototype.</figcaption></figure><h3>Generating an Ethereum Wallet</h3><p>The core feature of the wallet is generating an Ethereum account.</p><p>Using ethers.js, this is surprisingly simple.</p><pre>const wallet = ethers.Wallet.createRandom();</pre><p>This line generates a completely new wallet using secure random values.</p><p>The wallet object contains several properties, including:</p><ul><li>the wallet address</li><li>the private key</li></ul><p>The address is the public identifier that can receive cryptocurrency.</p><p>The private key is what proves ownership of the wallet and allows transactions to be signed.</p><p>For this prototype, I displayed both the wallet address and private key directly in the browser so I could see what was happening during testing.</p><p>Every time the page refreshes, the application generates a brand new wallet.</p><p>For example, the page might display something like:</p><p>Address: 0x...</p><p>Private Key: 0x...</p><p>Seeing this happen instantly made it clear how easily wallets can be generated programmatically.</p><p>Of course, real wallets don’t display private keys like this after creation. They normally encrypt them or store them securely.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_XrnNHKmaAJY_peHN5WN8w.png" /><figcaption>Code used to generate a new Ethereum wallet using the ethers.js library.</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zHY311lKqz2jiuJ3lkgs3w.png" /><figcaption>Example output showing a newly generated wallet address and private key.</figcaption></figure><h3>Connecting to the Ethereum Network</h3><p>After generating the wallet, the next step was connecting the application to the Ethereum network.</p><p>To read blockchain data, applications need something called a provider. A provider acts like a gateway that allows the program to communicate with the blockchain.</p><p>Using ethers.js, connecting to the network can be done with one line:</p><pre>const provider = ethers.getDefaultProvider(&quot;mainnet&quot;);</pre><p>This connects the application to the **Ethereum main network.</p><p>Once connected, the wallet balance can be fetched using:</p><pre>const balance = await provider.getBalance(wallet.address);</pre><p>However, Ethereum balances are returned in Wei, which is the smallest unit of ETH.</p><p>Wei values are extremely large numbers, so they need to be converted into ETH for easier reading.</p><pre>const ethBalance = ethers.formatEther(balance);</pre><p>After converting the balance, the application displays the wallet address, ETH balance, and private key in the browser.</p><p>At this point the wallet was actually pulling live data from the Ethereum blockchain, which was a pretty cool moment in the project.</p><h3>Challenges and Difficulties</h3><p>Even though the project itself isn’t huge, there were a few challenges along the way.</p><p>The first big issue happened right at the start when I tried using Replit and couldn’t easily find where the project files were. That ended up slowing things down until I switched to running the project locally.</p><p>Another challenge was understanding how private keys and wallet addresses are connected. At first, it wasn’t totally clear how a private key could generate an address or how that relationship worked.</p><p>Once I learned that the address is derived from cryptographic operations based on the private key, it started to make more sense.</p><p>Connecting to the blockchain also required figuring out how balances are returned in Wei and converting them to ETH, so they display properly.</p><p>Another small challenge was handling user input when importing wallets using private keys. If an invalid key is entered, the application needs to handle it safely instead of crashing.</p><p>Each of these problems helped me understand the underlying mechanics of blockchain development a bit better.</p><h3>What I Learned and Next Steps</h3><p>Building this wallet was a really good way to understand how crypto wallets actually work behind the scenes.</p><p>One of the biggest things I learned is that wallets don’t actually store cryptocurrency. They just manage the keys that control funds on the blockchain.</p><p>I also learned how powerful libraries like ethers.js are. With only a small amount of code, you can generate wallets, query blockchain data, and interact with the Ethereum network.</p><p>This project also helped me get more comfortable with Node.js, running servers locally, and connecting applications to external networks.</p><p>This version of the project is Replica 1.</p><p>Right now the wallet can:</p><ul><li>generate new wallets</li><li>import wallets using a private key</li><li>show the wallet address</li><li>display the ETH balance</li></ul><p>For Replica 2, I’m unsure of what to do but for the next step, maybe that could be adding transaction functionality so the wallet can send ETH to other addresses.</p><p>This project could grow into a more complete wallet with features like token balances, transaction history, and a better user interface.</p><p>For now, Replica 1 serves as a solid starting point for understanding how Ethereum wallets work at a technical level.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8b96ce51b748" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[When Inflation Breaks Trust: Stablecoins in Argentina’s Monetary Crisis]]></title>
            <link>https://medium.com/@tinatran._/when-inflation-breaks-trust-stablecoins-in-argentinas-monetary-crisis-2aa6255fd740?source=rss-29f1333fd966------2</link>
            <guid isPermaLink="false">https://medium.com/p/2aa6255fd740</guid>
            <category><![CDATA[economics]]></category>
            <category><![CDATA[argentina]]></category>
            <category><![CDATA[inflation]]></category>
            <category><![CDATA[stable-coin]]></category>
            <category><![CDATA[cryptocurrency]]></category>
            <dc:creator><![CDATA[Tina Tran]]></dc:creator>
            <pubDate>Thu, 26 Feb 2026 07:47:56 GMT</pubDate>
            <atom:updated>2026-02-26T07:48:24.901Z</atom:updated>
            <content:encoded><![CDATA[<h4>How Argentina’s inflation crisis is accelerating stablecoin adoption and reshaping monetary trust.</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2L3FLkoQlNAWZZREt5jpVQ.png" /><figcaption>Source: <a href="https://tradingeconomics.com/argentina/inflation-cpi">Argentina Inflation Rate</a></figcaption></figure><p>In 2023, Argentina’s inflation hit 211% annually. That doesn’t just mean prices went up, it meant your paycheck was losing value so fast that you had to question whether it would have the same buying power even a few weeks later. You get paid on Friday and rush to a currency exchange to convert your pesos into USD. Long lines form. People are anxious. The government sets monthly limits on how much foreign currency you can buy, pushing many to the black market, where exchange rates can be 1.5 to 3 times higher than the official rate. Argentina’s inflation crisis is rooted in years of fiscal deficits, currency controls, and monetary expansion, but for ordinary citizens, the result feels simple: their money loses value.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-pVvqYfYntiGD3ZZPjH_gA.png" /><figcaption>Official vs. Black Market USD/ARS Exchange Rate. Source: <a href="https://www.rethinkecon.it/argentina-the-bizarre-exchange-rate/">Argentina: The bizarre exchange rate — Rethinking Economics Italia</a></figcaption></figure><p>In recent years, another option has emerged, one that doesn’t require standing in long exchange lines or dealing with black market brokers. Stablecoins like USDT (Tether) and USDC (Circle) are digital tokens pegged to the US dollar, designed to maintain a 1:1 value. Crypto transaction volume in Argentina has reached roughly $91 billion in recent years, signaling that digital assets are not a fringe alternative. Anyone with a smartphone and internet access can hold them in a digital wallet instead of a local bank account. They can be transferred instantly across borders without handling physical cash or a middleman. Instead of rushing to a currency exchange on Friday, a young worker can convert pesos into digital dollars within minutes. Yet while stablecoins offer speed and accessibility, their growing role raises deeper questions about trust, sovereignty, and long-term monetary stability.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Dhh3QsF7FPCHTET0Smpkrw.png" /><figcaption>Latin America: Countries by Cryptocurrency Value Received, June 2022 — July 2023. Source: <a href="https://www.chainalysis.com/blog/latin-america-cryptocurrency-adoption/">Latin America Cryptocurrency Adoption: Data and Analysis</a></figcaption></figure><p>While stablecoins offer short-term protection against peso inflation, they rely heavily on trust in private issuers and US financial infrastructure. Most users cannot redeem directly with issuers, meaning liquidity depends on exchanges and market confidence. If panic spreads and large numbers attempt to exit at once, local markets may struggle to accommodate demand. At a national level, widespread adoption could push Argentina toward unofficial dollarization, weakening the peso further and reducing the central bank’s monetary control, even if the peso continues to coexist alongside stablecoins. In response, the government may attempt to impose stricter regulations or taxes, creating additional uncertainty.</p><p>However, if Argentinians can already buy physical USD, why turn to stablecoins at all? Access to dollars is often restricted by capital controls, shortages, and unfavorable black market rates. Holding large amounts of cash also carries security risks and makes cross-border transfers difficult. Stablecoins reduce these frictions by allowing users to store dollar-denominated value digitally and transfer it instantly through a smartphone. In a country where speed and access matter, digital dollars can feel more practical than paper ones.</p><p>When trust breaks, people don’t wait. They look for alternatives. Argentinians aren’t turning to stablecoins for fun, they’re turning to them because something deeper feels broken. When savings lose value and inflation surpassed 200% in 2023, trust in banks, central banks, and monetary policy begins to erode. Stablecoins are not the root solution; they are a response to instability and loss of trust. Ultimately, trust in a currency is not rebuilt through technology alone, but through consistent policy, credible institutions, and time. Until that trust is restored, alternatives like stablecoins will continue to fill the gap.</p><p>If inflation were to stabilize, some people might return to traditional banks. But trust, once broken, doesn’t fully reset overnight. People remember what it felt like to watch their purchasing power disappear.</p><p>This topic matters to me because I’ve seen how inflation affects real lives. During COVID, businesses shut down and uncertainty spread everywhere. Gas prices surged, and living in Vancouver, one of the most expensive cities in Canada, I’ve watched friends move away for affordable housing and families struggle with rising costs. Inflation isn’t just an economic statistic. It reshapes decisions, opportunities, and stability. Argentina may be an extreme case, but the underlying fear, losing control over your financial future is something people everywhere understand.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2aa6255fd740" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Using Augmented Reality to Make Biology Diagrams Easier to Understand]]></title>
            <link>https://medium.com/@tinatran._/using-augmented-reality-to-make-biology-diagrams-easier-to-understand-39c7b51a9579?source=rss-29f1333fd966------2</link>
            <guid isPermaLink="false">https://medium.com/p/39c7b51a9579</guid>
            <category><![CDATA[learning-design]]></category>
            <category><![CDATA[education]]></category>
            <category><![CDATA[augmented-reality]]></category>
            <category><![CDATA[students-projects]]></category>
            <category><![CDATA[technology]]></category>
            <dc:creator><![CDATA[Tina Tran]]></dc:creator>
            <pubDate>Tue, 16 Dec 2025 21:29:50 GMT</pubDate>
            <atom:updated>2025-12-16T21:29:50.615Z</atom:updated>
            <content:encoded><![CDATA[<p>from windshield arrows to textbook diagrams</p><p>When I was in grade 2, my dad was driving on the highway with the GPS on. Every few seconds, his eyes would move from the road down to the screen and back up again. I remember staring at the blue line on the GPS that told him when to switch lanes and thinking how strange it was that the instructions lived on a screen instead of where the driving was actually happening. I imagined those blue lines floating directly on the road, guiding the car without forcing him to look away. I did not have the language for it back then, but I was thinking about attention and overlays.</p><p>Years later in grade 10 biology, I noticed the same problem in a very different place. Complex processes like DNA structure, mitosis, and meiosis were explained using flat diagrams printed on paper. Teachers would tell us to study the diagrams, but most students skimmed them, flipped pages, or pulled out their phones to watch a video instead. The information was there, but it was not living where our attention naturally was. That gap between where the explanation lives and where your eyes are focused became the starting point for my first replicate project.</p><h3>The problem with textbook diagrams</h3><p>Biology diagrams try to explain systems that are three dimensional and dynamic using static two-dimensional images. DNA twists, chromosomes move, and cells change over time, but textbooks freeze these processes into a single snapshot. For a student, this creates a lot of invisible assumptions. you are expected to mentally rotate structures, imagine motion, and track sequences without any cues telling you what matters most.</p><p>Because of this, many students do not actually study diagrams deeply. They scan labels, glance at colors, and move on. When confusion hits, the usual response is to leave the textbook and search for a video online. that context switch has a real cost. Once a student opens their phone, notifications, messages, and other apps compete for attention. Even if they find a good video, by the time they return to the textbook, the mental thread is broken. Motivation drops, and the diagram becomes something to survive rather than understand.</p><p>This is not a lack of effort problem. It is a design problem. The explanation is separated from the object it is trying to explain. That separation is exactly where augmented reality can help.</p><h3>What I replicated</h3><p>For my first replicate, I built an augmented reality experience that adds video overlays directly on top of diagrams in my grade 10 biology textbook. Instead of leaving the page to search for an explanation, a student can scan the diagram with their phone and see an animation or explanation appear on top of the image itself.</p><p>I did not build an AR engine from scratch. My goal was to replicate an existing approach using Overlyapp, an image triggered AR platform that allows creators to attach digital content to physical images. Pranav, my TKS Director, recommended this tool to me after we scoped my original idea and realized that building a full custom AR system would take far longer than the time available for a first replicate.</p><p>From a user’s point of view, the experience is simple. You open the app, scan a diagram, and a video overlay appears explaining what is happening in that exact image. The complexity lives behind the scenes, but the interaction stays focused and minimal.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/660/1*gPoZgoLIX6TZTm66sq5xrQ@2x.jpeg" /></figure><p>I couldn’t get a video on here so heres a screenshot.</p><h3>How the system works at a high level</h3><p>At a high level, the system relies on image recognition rather than markers or QR codes. The app analyzes visual features in the diagram, such as edges and contrast patterns, and uses those features to recognize when the correct image is in view. Once the image is recognized, the app anchors a digital overlay to it so that the content appears fixed to the page.</p><p>The overlay itself does not need to be intelligent. The intelligence lives in the matching process. If the system can reliably recognize the diagram, it can reliably place content on top of it. Through testing, I learned that the system cares far more about visual features than how clean or realistic the image looks to a human. This difference between human perception and machine perception became very clear during the build process.</p><p>I did not know exactly how Overly implemented its recognition internally, but by experimenting with different images and settings, I could infer how sensitive it was to contrast and clarity.</p><h3>What broke when I tried to go beyond the demo</h3><p>The hardest and most educational part of this project was getting textbook diagrams to be recognized consistently. I initially uploaded around twenty photos of diagrams taken in good lighting. Despite this, the app repeatedly warned me that the images had low contrast and poor quality.</p><p>To fix this, I started heavily adjusting image settings. I boosted contrast, changed brightness, black point, highlights, shadows, and vibrance. The result looked terrible to a human. the images turned yellow, harsh, and unnatural. but the app stopped complaining, and recognition became more reliable.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*zDqYDz5oqhMRq6hg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*ltTLMoPMlP68IGOT" /></figure><p>This was my first real lesson in how computer vision systems see the world. Aesthetic quality does not matter as much as strong edges and clear feature separation. The image that looks worse to a person can look better to a machine. Putting the original diagram next to the over processed version makes this difference obvious.</p><p>I also attempted to use three dimensional models as overlays. I searched for higher quality DNA models on Sketchfab and tried converting files into formats supported by the platform. Even after resizing and simplifying the models, Overly rejected them as unsupported file types. Around the same time, my free trial ended, which limited me to video overlays only and capped the number of diagrams I could use.</p><p>This combination of technical limitations and product constraints forced me to rethink the project’s scope.</p><h3>Adapting instead of quitting</h3><p>At that point, it would have been easy to stop and call the project incomplete. Instead, I decided to adapt. I switched fully to video overlays and focused on making a small number of diagrams work well rather than trying to support everything.</p><p>This decision was influenced by something my mom has told me repeatedly. When something breaks, you do not freeze or quit. You adapt to the situation you are in and find a path forward. Finishing a constrained but functional project taught me more than abandoning a more ambitious one ever would have.</p><p>Video overlays also turned out to be a better fit for explaining motion and sequence, which mattered more than visual realism for this use case.</p><h3>How this changed how I think about AR</h3><p>Before this project, I thought augmented reality meant heavy coding and futuristic holograms projected onto glass. What I learned is that AR is often much simpler and more practical. It is about alignment, context, and attention. Placing the right information in the right place at the right time matters more than visual spectacle.</p><p>In this case, AR acted as a bridge between static content and dynamic explanation. It reduced context switching and kept learning anchored to the object of focus. That insight made AR feel less like science fiction and more like a design tool.</p><h3>Limitations and next steps</h3><p>This replicate has clear limitations. The free trial restricted the number of diagrams I could support, and the system relied on video rather than interactive three-dimensional models. The experience also depends on an external platform rather than custom infrastructure.</p><p>For a second replicate, my goal is much more ambitious. I want to build an AR app from scratch that covers the entire grade 10 biology textbook. Instead of relying on a no code platform, this version would involve writing actual code and supporting both video and 3d overlays across many diagrams. the goal would be to turn the textbook itself into an interactive surface, where students can scan any diagram and immediately see motion, structure, and explanation layered directly on top of what they are studying.</p><p>This first replicate was not about building something perfect. It was about understanding how augmented reality actually works in practice and how constraints shape real projects. That understanding is what sets up everything that comes next.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=39c7b51a9579" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why the Future of Driving Could Be in the Glass Itself: Exploring AR Windshields]]></title>
            <link>https://medium.com/@tinatran._/why-the-future-of-driving-could-be-in-the-glass-itself-exploring-ar-windshields-28c3234762dc?source=rss-29f1333fd966------2</link>
            <guid isPermaLink="false">https://medium.com/p/28c3234762dc</guid>
            <category><![CDATA[augmented-reality]]></category>
            <category><![CDATA[innovation]]></category>
            <category><![CDATA[cars]]></category>
            <category><![CDATA[future]]></category>
            <category><![CDATA[techonology]]></category>
            <dc:creator><![CDATA[Tina Tran]]></dc:creator>
            <pubDate>Tue, 04 Nov 2025 04:23:16 GMT</pubDate>
            <atom:updated>2025-11-04T04:23:16.126Z</atom:updated>
            <content:encoded><![CDATA[<h4>The glass in front of you might be the dashboard of the future.</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/603/1*uBeXQii8Xug0Vi_9Nz0poQ.png" /><figcaption>AR windshield displaying navigation and safety cues. Source: <a href="https://www.idropnews.com/news/apple-wants-to-make-a-futuristic-ar-windshield-for-autonomous-cars/79826/">idropnews.com</a></figcaption></figure><p>You’re driving at night, and the windshield can barely keep up with the rain. Every swipe spreads water across the glass, and the GPS keeps telling you to take turns you don’t even see. Your eyes jump between the dashboard and the road. The rain is so heavy the headlights can’t even help, barely lighting a few meters ahead. Your phone rings. Your dashboard flashes. In that split second, your attention slips. Your eyes are everywhere except the road. Driving shouldn’t feel this risky. What if your windshield could guide you, not just protect you from the rain?</p><h3><strong>HUDs</strong></h3><p>The thing is, it’s not the rain that is the problem. It’s split-second distractions. To fix this, carmakers make heads up displays (HUDs), small transparent screens on the lower windshield that show speed, alerts, and navigation. The goal is to keep the driver’s eyes on the road and lower distractions.</p><p>A projector shines light at an angle onto a mirror, which reflects it onto the windshield and into the driver’s eyes.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/778/1*GQrQWTBG8J4emDMAphNZoQ.png" /><figcaption>How heads-up display projects light onto the windshield. Source: <a href="https://www.embitel.com/blog/embedded-blog/how-combiner-glass-in-head-up-display-hud-works/">embitel.com</a></figcaption></figure><p>HUDs fix part of the problem, but they only show limited information in a small section of the glass. Drivers still need to look elsewhere for GPS or alerts, and light hitting the windshield layers wrong can cause a double image called ghosting. HUDs help reduce distractions, but with AR glass, the whole windshield becomes your dashboard. It becomes part of the drive. Imagine lane guidance, turn-by-turn directions, blind-spot warnings, and more. No more glancing; everything is in your line of sight.</p><h3>AR Windshields</h3><p>AR windshields are the next step from HUDs. They turn the whole glass into your guide, showing real-time instructions. Cameras, LiDAR, and radar track the surroundings, and AI mapping software figures out what’s nearby and how it’s moving. A clear OLED display projects cues onto the windshield. Lanes on the road get highlighted to remind you which one to take. Glowing arrows show your next turn. A pedestrian crossing ahead is outlined with soft light. When a car approaches your blind spot, the edge of the windshield flashes. Instead of looking down at screens and trying to piece all the information together, you have everything right in your line of sight.</p><h3>How They Work</h3><p>AR windshields combine multiple technologies. They mainly use three systems: sensors, AI processing, and display.</p><h4>LiDAR</h4><p>LiDAR sends out pulses of laser light and sees how long it takes to bounce off the surroundings. It uses that information to calculate how far away the object is. After it shoots millions of laser pulses, it forms a 3D map of its surroundings. It’s used for detecting pedestrians, vehicles, road edges, and obstacles.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yPd8WqBNrknUfQacaFKCAw.png" /><figcaption>LiDAR mapping its surroundings to form 3D view. Source: <a href="https://blogs.nvidia.com/blog/lidar-sensor-nvidia-drive/">blogs.nvidia.com</a></figcaption></figure><h4>Cameras</h4><p>Cameras act like the car’s eyes. They see how close a car is, lane markings, stop signs, etc. If a pedestrian is about to cross, the AR system outlines them and gives a small alert. At night, infrared cameras detect animals and humans before headlights see them.</p><h4>Radars</h4><p>Radars send out radio waves to see how long it takes to bounce off the object. They’re similar to LiDAR, but they calculate distance and speed — how far something is, how fast it’s moving, and if it is moving towards or away from you. They also send out longer wavelengths than LiDAR. This is helpful because it can sense a car slowing down ahead and alert you before you see the brake lights. The radio waves work in conditions that you can’t see, like fog or rain. They don’t see things like cameras, but they detect when something is there.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/836/1*z6dj9W6y8k1oTv2TmjtmcA.png" /><figcaption>Radar uses radio waves to measure speed and distance. Source: <a href="https://evsandbeyond.co.nz/radar-critical-for-mobilitys-future/">evsandbeyond.co.nz</a></figcaption></figure><h4>AI Processing</h4><p>AI processing is like the brain. It takes the data from LiDAR, cameras, and sensors. Every sensor sees things differently, but AI processing merges them into one. It predicts what each car, cyclist, and pedestrian is going to do. It filters only valuable information to be displayed on the windshield. They aren’t just TVs on glass; they combine the digital and physical world. To make the cues align, it tracks head movement, speed, and position. It tracks the driver’s head to align the display to the outside world.</p><h4>Display System</h4><p>After the system decides what to show on the glass, projection uses transparent OLEDs or waveguide projections. Keeping the cues aligned is the hardest part of AR technology. Even an error a little bit off can make the arrows or alerts show up in the wrong place. That’s why the system updates every millisecond using data from GPS and sensors to make sure the image stays where it belongs.</p><p>The challenges with this: bright sunlight can wash out the display, processing lag can cause things to display at the wrong time, and graphic alignment can be off. The cost of LiDAR and waveguide displays is expensive and only for a limited number of cars. Dirty windshields and harsh weather can distort the windshield display.</p><h3>The Road Ahead</h3><p>Driving is always about focus — keeping your eyes on the road and reacting when things go wrong. AR windshields are bringing that focus back, not by showing more, but by showing it in a smarter way. Instead of glancing between screens, everything you need to see is already on the glass. The question is, if the car sees everything for us, will we focus less? Technology can help us, but it can’t replace our awareness. The glass in front of us can do more than block rain. It could change how we see the road and maybe ourselves as drivers. As glass gets smarter, maybe the road ahead isn’t just about driving, it’s about learning to see better.</p><blockquote>Images are used for educational purposes to support explanations of AR windshield technology. All rights belong to their respective owners.</blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=28c3234762dc" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>