<?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 Simone Ruggiero on Medium]]></title>
        <description><![CDATA[Stories by Simone Ruggiero on Medium]]></description>
        <link>https://medium.com/@simonsruggi?source=rss-880181e9f4ec------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*hDtat7baQ0nGxzmqnEO8uw.jpeg</url>
            <title>Stories by Simone Ruggiero on Medium</title>
            <link>https://medium.com/@simonsruggi?source=rss-880181e9f4ec------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 02:36:01 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@simonsruggi/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[My MRR chart started on the day I installed my own app. So I fixed time.]]></title>
            <link>https://medium.com/@simonsruggi/my-mrr-chart-started-on-the-day-i-installed-my-own-app-so-i-fixed-time-dd0ca778158e?source=rss-880181e9f4ec------2</link>
            <guid isPermaLink="false">https://medium.com/p/dd0ca778158e</guid>
            <category><![CDATA[macos]]></category>
            <category><![CDATA[saas]]></category>
            <category><![CDATA[entrepreneurship]]></category>
            <category><![CDATA[open-source]]></category>
            <category><![CDATA[startup]]></category>
            <dc:creator><![CDATA[Simone Ruggiero]]></dc:creator>
            <pubDate>Tue, 27 Oct 2026 14:32:03 GMT</pubDate>
            <atom:updated>2026-10-27T14:32:03.307Z</atom:updated>
            <content:encoded><![CDATA[<p>Three small bugs in a menu bar MRR tracker, and what they taught me about charts that tell the truth.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/760/1*lqSOnQCb4ZagSuh3y_aoOA.png" /><figcaption>MRRDock with demo data</figcaption></figure><p><strong>MRRDock</strong> is a free, open-source macOS menu bar app that shows your MRR across Stripe, RevenueCat, Paddle, Lemon Squeezy, Polar, Dodo Payments, Gumroad and any custom endpoint — in one number, without an account, with your API keys in the macOS Keychain.</p><p>I launched it in September. Since then I’ve used it every day on my own apps, and the most useful thing daily use did was show me where the app was quietly lying. Here are three of those places.</p><h3>1. The chart started where MRRDock did, not where the business did</h3><p>MRRDock saves a reading of your MRR a few times a day and draws a chart from that history. That works — but only from the day you install it. Install it today and your chart is a single dot, and “growth over the last 30 days” has nothing to compare with.</p><p>The fix: some providers already keep their own MRR history. RevenueCat exposes it as a daily series, so <strong>MRRDock now backfills the chart from the provider once a day.</strong> Install it today and you see your real trend from the start. Providers that can’t answer that question keep the local readings.</p><p>And when there still isn’t enough history, MRRDock refuses to show a growth number at all. “+400% since Tuesday” is not information.</p><h3>2. A year of data gave today half the width</h3><p>The first chart plotted points by <strong>position in the array</strong>, not by time. Old readings had been compacted to one per day, while the last 48 hours kept every reading. Result: a few days of fine-grained data took the same horizontal space as months of daily points, and the chart looked like the business had exploded last week.</p><p>Now the x-axis is time. A year looks like a year, and there are proper ranges — 24H, 7D, 1Y and a custom range picked on a calendar. The growth pill follows the range you selected instead of always comparing with 30 days ago.</p><h3>3. The same account, counted twice</h3><p>RevenueCat project IDs show up in two forms: 1c0213f5 and proj1c0213f5. Paste one into MRRDock, later paste the other, and you had two sources pointing at the same project — and <strong>double the MRR</strong>.</p><p>It’s the kind of bug that makes you feel great for about four seconds. Now MRRDock accepts both forms, recognizes the same project (or the same custom endpoint URL) when it shows up twice, counts it once, marks the extra copy as DUPLICATE and refuses to save a new one.</p><h3>The lesson</h3><p>An MRR tracker has one job: give you a number you can trust at a glance. Every one of these bugs made the number or the chart <em>look</em> better than reality. That’s the dangerous direction: nobody files a bug when their revenue looks higher.</p><h3>Try it</h3><p>MRRDock is free and open source. No account, no server, no telemetry; keys live in the Keychain and the numbers never leave your Mac.</p><pre>brew install --cask simonsruggi/tap/mrrdock</pre><p>Or download it from <a href="https://github.com/simonsruggi/MRRDock/releases/latest">GitHub</a>.</p><p><strong>If it’s useful to you, please </strong><a href="https://github.com/simonsruggi/MRRDock"><strong>⭐ star it on GitHub</strong></a><strong>.</strong> Stars are the only signal I get, and they’re what gets an app into the official Homebrew catalog.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=dd0ca778158e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[I built a RAM monitor for the Mac menu bar. The hard part was making it not use RAM.]]></title>
            <link>https://medium.com/@simonsruggi/i-built-a-ram-monitor-for-the-mac-menu-bar-the-hard-part-was-making-it-not-use-ram-35f1cc966647?source=rss-880181e9f4ec------2</link>
            <guid isPermaLink="false">https://medium.com/p/35f1cc966647</guid>
            <category><![CDATA[macos]]></category>
            <category><![CDATA[open-source]]></category>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[swift]]></category>
            <category><![CDATA[productivity]]></category>
            <dc:creator><![CDATA[Simone Ruggiero]]></dc:creator>
            <pubDate>Tue, 27 Oct 2026 14:30:38 GMT</pubDate>
            <atom:updated>2026-10-27T14:30:38.364Z</atom:updated>
            <content:encoded><![CDATA[<p>An empty NSPopover costs 25 MB. Here’s what I used instead, and how RamDock idles at ~13 MB.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/972/1*r2ZqZrOWmSk8oFI7XIU9SA.png" /></figure><p>When the fans spin up or the Mac starts to stutter, the question is always the same: <em>what is eating my RAM or my CPU right now, and can I kill it?</em></p><p>Activity Monitor answers it, but it’s a full window, a dozen columns and three clicks away. I wanted the answer in the menu bar: two numbers, always visible, and one click to the process that’s causing trouble.</p><p>So I built <strong>RamDock</strong> — a free, open-source RAM and CPU monitor for the macOS menu bar.</p><h3>What it does</h3><ul><li><strong>Two numbers, always visible.</strong> System RAM and CPU usage, refreshed every 3 seconds, colored from green to yellow to red as the load rises.</li><li><strong>One click to the culprit.</strong> Every process you own, with its CPU and memory, sortable by either, searchable by name or PID.</li><li><strong>One more click to fix it.</strong> The ✕ on each row quits the process. ⌥-click or right-click → <em>Force Quit</em> if it doesn’t listen.</li><li><strong>A “This Mac” panel</strong> with model, chip, CPU and GPU cores, memory, storage, macOS version and uptime.</li></ul><h3>The rule: a monitor must not become the problem</h3><p>A resource monitor that eats resources defeats its purpose. So I measured everything with footprint, and three decisions came out of it.</p><h3>1. No NSPopover</h3><p>The obvious way to build a menu bar dropdown on macOS is NSPopover. I started there. Then I measured it: <strong>an </strong><strong>NSPopover keeps about 25 MB alive after its first open — even with an empty view.</strong></p><p>A translucent borderless NSPanel with real controls in it costs about 18 MB, and unlike the popover it can be thrown away completely. RamDock builds the panel when you click and releases it when you close it.</p><p>There was a second reason. A closed NSPopover keeps its SwiftUI view alive, so onDisappear never fired and the process sampler kept running in the background forever. With a panel that&#39;s destroyed on close, there&#39;s nothing left to keep running.</p><h3>2. Sample processes only when someone is looking</h3><p>Listing every process with its CPU and memory isn’t free: it’s proc_listallpids plus proc_pid_rusage for each PID, every 2 seconds. RamDock does that <strong>only while the panel is open</strong>. Closing the panel stops sampling and releases the list.</p><p>The menu bar readout itself reads two kernel counters every 3 seconds and redraws only when a number actually changes.</p><h3>3. Plain AppKit</h3><p>RamDock is written in AppKit, not wrapped in a web view or a cross-platform framework. The whole app is small enough that this is easy, and it shows in the numbers.</p><h3>The numbers</h3><p>Measured with footprint on an M5 MacBook Pro:</p><ul><li><strong>~13 MB</strong> of memory at launch, <strong>~27–35 MB</strong> after using the panel</li><li><strong>~0.15 s of CPU per minute</strong> with the panel closed</li></ul><p>The memory figures are the same ones Activity Monitor shows in its “Memory” column (ri_phys_footprint), so you can check them yourself.</p><h3>Get it</h3><p>RamDock is free, open source (MIT), and has no account and no telemetry. It updates itself through Sparkle.</p><pre>brew install --cask simonsruggi/tap/ramdock</pre><p>Or grab the latest release from <a href="https://github.com/simonsruggi/RamDock/releases/latest">GitHub</a>.</p><p><strong>If you find it useful, a </strong><a href="https://github.com/simonsruggi/RamDock"><strong>⭐ star on GitHub</strong></a><strong> helps more than you’d think.</strong> It’s the only signal I get that people are using it, and it’s what gets an app into the official Homebrew catalog.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=35f1cc966647" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[My menu bar app was using 8.5 GB of RAM. Here’s the bug.]]></title>
            <link>https://medium.com/@simonsruggi/my-menu-bar-app-was-using-8-5-gb-of-ram-heres-the-bug-e6bb9fa9a234?source=rss-880181e9f4ec------2</link>
            <guid isPermaLink="false">https://medium.com/p/e6bb9fa9a234</guid>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[swift]]></category>
            <category><![CDATA[open-source]]></category>
            <category><![CDATA[debugging]]></category>
            <category><![CDATA[macos]]></category>
            <dc:creator><![CDATA[Simone Ruggiero]]></dc:creator>
            <pubDate>Tue, 27 Oct 2026 14:29:04 GMT</pubDate>
            <atom:updated>2026-10-27T14:29:04.764Z</atom:updated>
            <content:encoded><![CDATA[<p>A WebSocket that reconnected every 3 seconds for three and a half days — and why macOS never let go of the old ones.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*iDoNebNInsspZDbBGD7Q2A.png" /></figure><p>StockDock is a free, open-source macOS menu bar app that shows your stocks, crypto and portfolio P&amp;L in real time. It’s supposed to be the lightest thing running on your Mac. So when I opened Activity Monitor and saw it sitting at <strong>8.5 GB of memory after 3.5 days of uptime</strong>, I took it personally.</p><p>This is the story of that leak: what caused it, how I found it with two built-in macOS command-line tools, and the two small changes that fixed it.</p><h3>The feature that started it</h3><p>A few days earlier I had shipped <strong>overnight and weekend prices</strong>. US stocks stop trading at 8 PM New York time and don’t start again until 4 AM, and on weekends they don’t trade at all. But Hyperliquid lists 24/7 perpetual futures that track many US stocks, so StockDock can now show a “24/7” price when the exchange is closed.</p><p>The implementation was simple: open a WebSocket to Hyperliquid, subscribe to the coins that match your watchlist, update the price on every tick. If the socket drops, reconnect with exponential backoff.</p><p>That last sentence is where the bug lived.</p><h3>Measuring instead of guessing</h3><p>Activity Monitor tells you <em>that</em> memory is growing. It doesn’t tell you <em>what</em> it is. macOS ships two tools that do:</p><pre>footprint &lt;pid&gt;<br>heap &lt;pid&gt; --sortBySize</pre><p>footprint gives the real memory footprint of a process, broken down by category. heap walks the malloc zones and groups live objects by class. Sorted by size, the top of the list was unambiguous: __NSURLSessionWebSocketTask, counted in the thousands.</p><p>The app should have had one WebSocket. It was holding on to every one it had ever opened.</p><h3>The bug</h3><p>Two things combined:</p><ol><li><strong>Hyperliquid closes the socket if you subscribe to a coin it doesn’t list.</strong> If your watchlist contains a stock with no 24/7 market, the subscription request makes the server drop the connection.</li><li><strong>My backoff was reset as soon as the subscription was confirmed.</strong> The idea was “the connection is healthy again, start the next retry from the shortest delay.” But the confirmation arrived <em>before</em> the server got to the unlisted coin and closed the socket.</li></ol><p>So the loop was: connect → subscription confirmed → backoff reset → server closes the socket → reconnect after the shortest delay → repeat. <strong>Roughly every 3 seconds, forever.</strong></p><p>On its own, a reconnect loop wastes battery and network, but it shouldn’t leak. The second half of the story is that each failed URLSessionWebSocketTask stayed retained by the system — around <strong>80 KB each</strong>. At one reconnect every 3 seconds, 3.5 days is about 100,000 reconnects. 100,000 × 80 KB ≈ 8 GB. The numbers matched.</p><h3>The fix</h3><p>Two changes, both small:</p><ul><li><strong>Only subscribe to coins Hyperliquid actually lists.</strong> StockDock now keeps the list of active coins and never asks for one that isn’t there, so the server has no reason to close the socket.</li><li><strong>Reset the backoff only when a real price arrives</strong>, not when the subscription is acknowledged. An acknowledgement proves the server heard you; a price proves the connection works. If something else ever makes the socket drop right after connecting, the delay now grows instead of hammering the server every few seconds.</li></ul><p>After the fix, StockDock runs for days and stays where it should be.</p><h3>What I’m taking away from it</h3><ul><li><strong>“Connected” is not “working.”</strong> Reset your retry state on the first piece of real data, not on the handshake.</li><li><strong>Leaks in long-running apps hide in the slow paths.</strong> This one was invisible in a 10-minute test. It only showed up because a menu bar app is meant to run for weeks.</li><li><strong>heap --sortBySize is underrated.</strong> One command, and the class name told me exactly where to look.</li></ul><h3>Try it</h3><p>StockDock is free, open source and has no account, no API keys and no tracking. It shows your watchlist and portfolio in the menu bar and in a full desktop window, with pre-market, after-hours and now 24/7 prices.</p><pre>brew install --cask simonsruggi/tap/stockdock</pre><p>Or download it from <a href="https://stockdockapp.com/">stockdockapp.com</a>.</p><p><strong>If you use it, please </strong><a href="https://github.com/simonsruggi/StockDock"><strong>⭐ star the repo on GitHub</strong></a><strong>.</strong> There’s no tracking in the app, so a star is the only way I know anyone is out there — and enough of them would let StockDock into the official Homebrew catalog.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e6bb9fa9a234" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[I built a free macOS menu bar app that shows my MRR across every payment platform]]></title>
            <link>https://medium.com/@simonsruggi/i-built-a-free-macos-menu-bar-app-that-shows-my-mrr-across-every-payment-platform-78997b76a2f3?source=rss-880181e9f4ec------2</link>
            <guid isPermaLink="false">https://medium.com/p/78997b76a2f3</guid>
            <category><![CDATA[open-source]]></category>
            <category><![CDATA[swiftui]]></category>
            <category><![CDATA[indie-hacking]]></category>
            <category><![CDATA[saas]]></category>
            <category><![CDATA[macos]]></category>
            <dc:creator><![CDATA[Simone Ruggiero]]></dc:creator>
            <pubDate>Mon, 07 Sep 2026 09:37:35 GMT</pubDate>
            <atom:updated>2026-09-07T09:37:35.474Z</atom:updated>
            <content:encoded><![CDATA[<h4>Stripe, RevenueCat, Paddle, Lemon Squeezy, Polar, Dodo and Gumroad, added up every fifteen minutes. No account, no server, MIT licensed.</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/760/1*lqSOnQCb4ZagSuh3y_aoOA.png" /><figcaption>MRRDock’s popover: total MRR, 30-day change and the per-source breakdown. Figures are demo data.</figcaption></figure><p>I sell software in more places than I’d like to admit.</p><p>The iOS apps report through RevenueCat. A web product bills on Stripe. Paddle handles a desktop app because it’s a merchant of record and VAT is not a hobby I want. Templates sit on Lemon Squeezy. Three of those bill in dollars, I think in euros.</p><p>So the question “how am I actually doing this month?” costs me four logins, four dashboards, and a mental sum with a currency conversion in the middle of it. I was asking that question roughly twenty times a day, and answering it badly.</p><p>MRRDock is the answer to that question, sitting in the menu bar:</p><pre>brew install --cask simonsruggi/tap/mrrdock</pre><p>One number. All platforms. Your currency. Free, open source, and the data never leaves your Mac.</p><h3>What it does</h3><p>It’s a menu bar app — no Dock icon, no window. You add your sources once, and from then on the total MRR is always visible. Click it and the popover breaks the number down.</p><ul><li><strong>Multi-platform aggregation.</strong> Stripe, RevenueCat, Paddle Billing, Lemon Squeezy, Polar, Dodo Payments, Gumroad, plus any HTTPS endpoint of your own. Multiple accounts of the same platform are fine: two Stripe accounts and four RevenueCat projects add up like anything else.</li><li><strong>Multi-currency.</strong> Every source converts into your display currency with live rates cached for 12 hours. Amounts that can’t be converted are shown separately, never silently folded into the total.</li><li><strong>A real MRR trend.</strong> 24H, 7D, 1Y or a custom range, with a “vs 30 days ago” delta.</li><li><strong>Seven menu bar modes.</strong> MRR, MRR + 30-day change, ARR, 28-day revenue, active subscriptions, one source at a time cycling every four seconds, or icon only.</li><li><strong>Privacy mode.</strong> One click replaces every figure with ••• for screen sharing or a café.</li><li><strong>Discord and Slack notifications.</strong> MRR milestones (“just crossed €5,000”) and an optional daily summary.</li></ul><h3>The part I care about: honest arithmetic</h3><p>Anyone can call four APIs and add up the numbers. The interesting work is in deciding what a number <em>means</em>, and being loud about the cases where you don’t know.</p><p>Every recurring line item gets priced the same way:</p><pre>monthly = unit_price × quantity × (intervals_per_month ÷ interval_count) × (1 − discount)</pre><p>intervals_per_month is <strong>30.4375</strong> for daily, <strong>4.348</strong> for weekly, <strong>1</strong> for monthly and <strong>1/12</strong> for yearly. That&#39;s the average Gregorian month — 365.25 ÷ 12 — chosen so that twelve monthly plans and one yearly plan reconcile to the cent instead of quietly drifting apart over a year.</p><p>Then there are the judgement calls. These are the ones that make any two dashboards disagree, so MRRDock documents all of them:</p><ul><li><strong>Free trial</strong> — counted in Trials, contributes €0 to MRR.</li><li><strong>Yearly plan</strong> — divided by 12. <strong>Seats / quantity</strong> — multiplied.</li><li><strong>Percentage coupon (−20%)</strong> — applied.</li><li><strong>Fixed-amount coupon (−€5)</strong> — <em>not</em> applied: prorating it across a multi-item subscription needs data these APIs don’t return.</li><li><strong>Usage-based / metered price</strong> — <em>not guessed.</em> Flagged as “could not be priced” so the understatement is visible.</li><li><strong>Past due / grace period</strong> — follows the provider’s own status.</li><li><strong>ARR</strong> — exactly MRR × 12, nothing smoothed behind your back.</li></ul><p>The rule I kept coming back to: <strong>an understated total you can see beats a confident total that’s wrong.</strong> A metered plan MRRDock can’t price shows up as a warning in the popover, not as a zero it hopes you won’t notice.</p><p>Gumroad gets the same treatment at the source level. Its API exposes sales, not a subscription ledger you can price — so deriving an MRR from one-off sales would be a guess. That source contributes 28-day revenue only, and the UI says exactly that.</p><h3>When my app disagreed with RevenueCat, and both were right</h3><p>Two days after I started using it, MRRDock said 303.76 € across four RevenueCat projects. The RevenueCat dashboard said 306 €. It also said 11 active trials against my 9.</p><p>The instinct is to assume you have a bug. I spent an evening proving I didn’t, and the answer is a nice illustration of why aggregation is harder than it looks:</p><ol><li><strong>/v2/metrics/overview returns MRR rounded to whole dollars.</strong> Four projects reporting 181, 127, 27 and 18 dollars carry up to ±2 $ of rounding — about ±1.7 € once converted. That’s most of the gap right there.</li><li><strong>We use different exchange rates.</strong> Mine come from Yahoo, cached for 12 hours. RevenueCat uses its own, at its own moment.</li><li><strong>The trials were mine.</strong> Four RevenueCat projects were sitting at zero MRR so I never bothered adding them to MRRDock — but one of them had 2 live trials. 9 + 2 = 11. Not a bug: an incomplete configuration, which the per-source breakdown made obvious in about ten seconds.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/760/1*C1BMjJulGSi_CXa1Xy_0dA.png" /><figcaption>The per-source breakdown: what each platform contributes and its share of the total. Demo data.</figcaption></figure><p>Expected residual drift: around 0.7%. Neither number is wrong. That’s worth knowing <em>before</em> you go looking for a bug in your own code at midnight.</p><h3>The chart should show the business, not the install date</h3><p>The first version stored a daily snapshot locally and drew a sparkline from it. Which means a brand-new install shows a flat line, and your MRR history “starts” the day you discovered the app. Useless.</p><p>So MRRDock backfills from the provider’s own daily MRR series. RevenueCat exposes it:</p><pre>GET /v2/projects/{project}/charts/mrr?resolution=day<br>    &amp;start_date=YYYY-MM-DD&amp;end_date=YYYY-MM-DD</pre><p>Same series the dashboard draws. Two things cost me time and are worth writing down: the dates <strong>must</strong> be YYYY-MM-DD (epoch timestamps get a 400), and outside the range where data exists the API answers with zeros rather than an empty array — so you have to trim the leading zeros yourself or your chart opens with a cliff that never happened. The last point comes back with incomplete: true, meaning today-so-far: discard it.</p><p>Providers that can’t answer keep their locally recorded readings. You get the longest honest history each source can give you.</p><h3>Your keys, your Mac</h3><p>There is no MRRDock backend. There is nothing to sign up for. I could not see your revenue if I wanted to.</p><ul><li>API keys live in the <strong>macOS Keychain</strong>, never in the app’s JSON file, never in logs, never in a config you might paste into a GitHub issue.</li><li>Requests go only to the platforms you configure, plus Yahoo Finance for exchange rates — a currency pair, nothing about you.</li><li>No account, no login, no analytics, no telemetry, no crash reporting.</li><li>HTTPS only: a custom endpoint on http:// is refused rather than handed your token in clear.</li><li><strong>Read-only by construction.</strong> The app issues GET requests and nothing else. A key with write scope still can’t change anything through MRRDock.</li></ul><p>That last point is why the <em>custom endpoint</em> source exists, incidentally. Some platforms have no public revenue API at all — Superwall, App Store Connect. Point MRRDock at any HTTPS URL of yours that returns the numbers, and it becomes a source like the others.</p><h3>What it isn’t</h3><p>It’s not a replacement for ChartMogul, Baremetrics or the RevenueCat dashboard. Cohorts, LTV, churn analysis — those belong on a real screen, and I’m not going to do them worse inside a popover.</p><p>MRRDock answers the one question you ask twenty times a day, without opening a browser.</p><h3>Try it</h3><p>Native SwiftUI, no third-party dependencies, macOS 14+, universal binary, signed with a Developer ID and notarized by Apple, with built-in auto-updates. MIT licensed.</p><pre>brew install --cask simonsruggi/tap/mrrdock</pre><p>Or grab the zip from the <a href="https://github.com/simonsruggi/MRRDock/releases/latest">releases page</a>.</p><p>The code is on GitHub: <a href="https://github.com/simonsruggi/MRRDock">github.com/simonsruggi/MRRDock</a>. Issues and pull requests welcome — especially if you sell through a platform I haven’t covered yet.</p><p>If the number goes up, that’s on you. I just made it easier to look at.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=78997b76a2f3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[StockDock 1.9.0: My Free, Open-Source Stock Tracker for Mac Just Became a Full Desktop App]]></title>
            <link>https://medium.com/@simonsruggi/stockdock-1-9-0-my-free-open-source-stock-tracker-for-mac-just-became-a-full-desktop-app-4cc09df71e5b?source=rss-880181e9f4ec------2</link>
            <guid isPermaLink="false">https://medium.com/p/4cc09df71e5b</guid>
            <category><![CDATA[stocks]]></category>
            <category><![CDATA[open-source]]></category>
            <category><![CDATA[swift]]></category>
            <category><![CDATA[indie-development]]></category>
            <category><![CDATA[macos]]></category>
            <dc:creator><![CDATA[Simone Ruggiero]]></dc:creator>
            <pubDate>Thu, 16 Jul 2026 12:27:07 GMT</pubDate>
            <atom:updated>2026-07-16T12:27:07.614Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*d5Fde6VZrhf08iUw6p5OUg.png" /><figcaption>The new StockDock desktop window — Home, Watchlist, Portfolios and Settings in one view, in sync with the menu bar.</figcaption></figure><p>A year ago I got tired of opening a browser tab every time I wanted to check my portfolio. So I built <strong>StockDock</strong> — a tiny macOS menu bar app that shows your stocks, crypto, and P&amp;L in real time, one click away, with no account and no subscription.</p><p>It’s stayed <strong>100% free and open source</strong> the whole way. And with the release I just shipped — <strong>1.9.0</strong> — it grew up: StockDock is now a full desktop app <em>and</em> a menu bar app at the same time.</p><p>Let me show you what’s new, and why “free and open source, forever” is a promise I actually mean.</p><h3>The one-line pitch</h3><p><strong>StockDock is a free, open-source macOS app for tracking stocks, ETFs, indices, crypto, and your portfolio P&amp;L in real time — from your menu bar or a full desktop window.</strong></p><p>No account. No API keys. No subscription. No tracking. Your watchlist and portfolio never leave your Mac.</p><p>brew install simonsruggi/tap/stockdock</p><h3>What’s new in 1.9.0 — a full desktop window</h3><p>Until now, StockDock lived entirely in your menu bar: click the icon, a popover drops down, done. That’s still there — but sometimes you want room to breathe. So now the popover has an <strong>Open</strong> button that expands StockDock into the complete desktop window above: News, Watchlist, Portfolios, and Settings in one spacious, redesigned view — with live charts, allocation breakdowns, and your day’s movers.</p><p>Everything stays perfectly in sync with the menu bar: change something in the window and the popover updates instantly, and vice versa. Same app, same data, two ways to look at it.</p><p>Also new in 1.9.0:</p><ul><li><strong>Price &amp; After-hours side by side</strong> — separate, individually sortable columns for the regular price and the pre/post-market price (each with its own % move).</li><li><strong>News search</strong> — filter stories by headline, ticker, or publisher.</li><li><strong>Add to portfolio from a stock’s detail</strong> — open any symbol and add it as a position in one click.</li><li><strong>Sharper precision</strong> — forex pairs and sub-dollar prices now show the decimals that actually matter (0.7119, not 0.71).</li><li><strong>Decimal controls</strong> — choose how many decimals to show for percentages and values, or hide the percentage in the menu bar entirely.</li></ul><h3>The problem I was solving</h3><p>Most ways to watch the market fall into two camps:</p><ol><li><strong>Full trading platforms</strong> (Robinhood, IBKR, Trading 212) — great for trading, overkill for glancing. They take over the screen and drain the battery.</li><li><strong>Web dashboards</strong> (Yahoo Finance, Google Finance) — a browser tab you have to keep open, buried in ads and distractions.</li></ol><p>I wanted the in-between: a quiet widget that just shows the numbers. Like the battery percentage or Wi-Fi status in your menu bar — but for your portfolio. StockDock is that. And now, when you <em>do</em> want the full picture, it’s one click away.</p><h3>What you can track</h3><p>Stocks, ETFs, indices (S&amp;P 500, NASDAQ…), crypto, and forex — anything with a Yahoo Finance symbol. Search by <strong>symbol</strong>, <strong>company name</strong>, or <strong>ISIN</strong> (AAPL, Tesla, IE00B4L5Y983).</p><p>A few things people tend to like:</p><ul><li><strong>Menu bar P&amp;L</strong> — pick what shows up top: P&amp;L, total value, % change, best/worst mover, or just an icon. Green when you’re up, red when you’re down.</li><li><strong>Multi-currency portfolios</strong> — hold mixed-currency positions and see totals converted to yours, with the historical exchange rate applied to your cost basis so P&amp;L stays honest.</li><li><strong>Long/short + leverage</strong> — optional, off by default, with correct sign and exposure math.</li><li><strong>Alerts that reach you</strong> — one-shot price alerts and per-portfolio notifications, delivered as native macOS notifications and optionally mirrored to a Discord or Slack webhook.</li><li><strong>52-week range bars, extended-hours prices, export/import to JSON</strong> — the details that make it feel finished.</li></ul><h3>Private by design (and I mean it)</h3><p>This part matters to me more than any feature:</p><ul><li>Everything is stored locally on your Mac.</li><li>The app only ever talks to Yahoo Finance’s public endpoints (and your own webhook, if you set one up).</li><li><strong>No account. No login. No analytics. No telemetry.</strong> I have literally no way of knowing how many people use StockDock.</li><li>It’s fully open source — you can read every line before you trust it with your watchlist.</li></ul><p>In a world where every app wants your email, StockDock wants nothing.</p><h3>Why open source, and how you can help</h3><p>StockDock is MIT-licensed and it will <strong>always</strong> be free — no paid tier, no premium lock, no “unlock pro” wall. That’s a deliberate choice, not a soft launch for a subscription later.</p><p>But that same privacy stance means I’m flying blind: no analytics means the only signal I ever get that the app is worth maintaining is a <strong>GitHub star</strong>. So if StockDock is useful to you, the single most helpful thing you can do — and it’s free — is <a href="https://github.com/simonsruggi/StockDock">star the repo</a>.</p><p>And if you want to go further, you can <a href="https://github.com/sponsors/simonsruggi">sponsor me on GitHub</a>. It changes nothing about the app — every feature stays free for everyone — it just helps me keep building. Either way, thank you.</p><h3>Install</h3><p><strong>Homebrew</strong> (recommended): brew install simonsruggi/tap/stockdock</p><p>Or grab the latest StockDock.zip from <a href="https://github.com/simonsruggi/StockDock/releases/latest">GitHub Releases</a>, unzip, and drop it in /Applications.</p><p>The app is code-signed, notarized by Apple, and auto-updates via Sparkle — install once and forget about it. Requires <strong>macOS 14 Sonoma or later</strong>, and runs natively on both Apple Silicon and Intel.</p><h3>For the nerds</h3><ul><li><strong>Swift + SwiftUI</strong> — native macOS, no Electron, no web views.</li><li><strong>Yahoo Finance WebSocket</strong> — real-time prices (~1 tick/second), decoded from binary with swift-protobuf.</li><li><strong>Sparkle</strong> — seamless background auto-updates.</li><li><strong>Universal binary</strong> — ARM64 + x86_64 in one small, fast app.</li></ul><p>Six languages, too: English, German, French, Spanish, Italian, and Portuguese.</p><h3>Try it</h3><p>If you invest or trade and you’re on a Mac, StockDock takes about ten seconds to install and you might never close it again.</p><ul><li><strong>GitHub:</strong> <a href="https://github.com/simonsruggi/StockDock">github.com/simonsruggi/StockDock</a></li><li><strong>Install:</strong> brew install simonsruggi/tap/stockdock</li><li><strong>Star it</strong> if it’s useful — it genuinely helps me keep going.</li></ul><p><em>I’m Simone Ruggiero, an indie developer building apps for iOS and macOS. More of my work at </em><a href="https://simoneruggiero.com/"><em>simoneruggiero.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4cc09df71e5b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[I Replaced My Trading App With a 4MB Menu Bar Widget]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@simonsruggi/i-replaced-my-trading-app-with-a-4mb-menu-bar-widget-8d7f605d39f5?source=rss-880181e9f4ec------2"><img src="https://cdn-images-1.medium.com/max/2400/1*iDoNebNInsspZDbBGD7Q2A.png" width="2400"></a></p><p class="medium-feed-snippet"># I Built a Free macOS Menu Bar App to Track Stocks &#x2014; Here&#x2019;s Why</p><p class="medium-feed-link"><a href="https://medium.com/@simonsruggi/i-replaced-my-trading-app-with-a-4mb-menu-bar-widget-8d7f605d39f5?source=rss-880181e9f4ec------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@simonsruggi/i-replaced-my-trading-app-with-a-4mb-menu-bar-widget-8d7f605d39f5?source=rss-880181e9f4ec------2</link>
            <guid isPermaLink="false">https://medium.com/p/8d7f605d39f5</guid>
            <category><![CDATA[macos]]></category>
            <category><![CDATA[swiftui]]></category>
            <category><![CDATA[stock-market]]></category>
            <category><![CDATA[micron]]></category>
            <category><![CDATA[open-source]]></category>
            <dc:creator><![CDATA[Simone Ruggiero]]></dc:creator>
            <pubDate>Tue, 26 May 2026 16:14:16 GMT</pubDate>
            <atom:updated>2026-05-26T16:14:16.542Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[No More App Rejection: The Only iOS App Launch Checklist You'll Ever Need]]></title>
            <link>https://medium.com/@simonsruggi/no-more-app-rejection-the-only-ios-app-launch-checklist-youll-ever-need-d16d1645e7da?source=rss-880181e9f4ec------2</link>
            <guid isPermaLink="false">https://medium.com/p/d16d1645e7da</guid>
            <category><![CDATA[ios-app-development]]></category>
            <category><![CDATA[app-store]]></category>
            <category><![CDATA[ios-development]]></category>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[apple]]></category>
            <dc:creator><![CDATA[Simone Ruggiero]]></dc:creator>
            <pubDate>Tue, 17 Mar 2026 21:47:28 GMT</pubDate>
            <atom:updated>2026-03-17T21:47:28.417Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*efrNXp80T31J6htkKAjKTw.png" /></figure><p>Getting your app rejected by Apple is painful — and almost always avoidable. After shipping multiple iOS apps, I put together this checklist of everything that must be done before you hit &quot;Submit for Review.&quot; Follow it to the letter and you&#39;ll get approved on the first try.</p><blockquote>1. Apple Developer Account</blockquote><pre>Enrolled in the **Apple Developer Program** ($99/year) at [developer.apple.com](https://developer.apple.com)</pre><pre>Your membership is **active** (not expired — check the renewal date)</pre><pre>Two-factor authentication is enabled on your Apple ID</pre><pre>If shipping under a company: **D-U-N-S number** obtained and entity name matches App Store Connect</pre><blockquote>2. App Store Connect Setup</blockquote><pre>New app created in [App Store Connect](https://appstoreconnect.apple.com)</pre><pre>**Bundle ID** registered in Certificates, Identifiers &amp; Profiles — matches exactly what&#39;s in Xcode</pre><pre>Correct **primary language** selected</pre><pre>App name is unique (search the App Store before committing)</pre><pre>**SKU** set (internal identifier, not visible to users)</pre><pre>All **required roles** assigned if working in a team (Admin, Developer, etc.)</pre><blockquote>3. Xcode Project Configuration</blockquote><pre>`CFBundleIdentifier` matches the registered Bundle ID exactly</pre><pre>`CFBundleShortVersionString` (Marketing Version) set correctly — e.g. `1.0.0`</pre><pre>`CFBundleVersion` (Build Number) is a positive integer and **higher than any previous build** uploaded to App Store Connect</pre><pre>Deployment target set to a **supported iOS version** (Apple rejects apps targeting versions below their minimum)</pre><pre>All **capabilities** enabled in Xcode match those enabled in the App ID on the Developer Portal (Push Notifications, Sign in with Apple, In-App Purchases, etc.)</pre><pre>**No hardcoded IP addresses or localhost URLs** in production builds</pre><pre>`DEBUG` flags and test credentials are **not included** in the release build</pre><pre>Unused permissions are **removed** from `Info.plist`</pre><pre>Every permission used has a **human-readable usage description** in `Info.plist`:</pre><p>- NSCameraUsageDescription</p><p>- NSPhotoLibraryUsageDescription</p><p>- NSLocationWhenInUseUsageDescription</p><p>- NSMicrophoneUsageDescription</p><p>- NSContactsUsageDescription</p><p>- (add any others your app requires)</p><pre>`NSAppTransportSecurity` exceptions are justified — no blanket `NSAllowsArbitraryLoads: YES` in production</pre><p><em>Built with the </em><strong><em>latest stable Xcode version</em></strong><em> — Apple rejects apps built with outdated SDKs (currently Xcode 16+)</em></p><blockquote>4. Code Signing &amp; Provisioning</blockquote><pre>**Distribution certificate** created and installed in Keychain</pre><pre>**App Store provisioning profile** created for the Bundle ID and linked to the distribution certificate</pre><pre>In Xcode Build Settings: Signing set to **Release** configuration with the correct profile</pre><pre>`Automatically manage signing` is either fully trusted or fully manual — do not mix</pre><pre>Archive builds without warnings in the signing section</pre><blockquote>5. App Icons</blockquote><pre>Icon provided in all **required sizes** (use an asset catalog `AppIcon` in Xcode)</pre><pre>**1024×1024 px** App Store icon included (no alpha channel, no rounded corners — Apple rounds it)</pre><pre>Icon has **no transparency** on any size</pre><pre>Icon does not contain the Apple logo or any third-party IP without permission</pre><pre>Icon looks good when scaled down — Apple auto-generates all sizes from the 1024×1024 source</pre><blockquote>6. Launch Screen</blockquote><pre>Launch screen configured (Storyboard or `UILaunchScreen` key in `Info.plist`)</pre><pre>Launch screen does **not** contain any text that implies the app is loading a specific feature</pre><pre>Supports all target screen sizes (test on iPhone SE, iPhone 15 Pro Max, iPad if universal)</pre><blockquote>7. App Metadata (App Store Connect)</blockquote><h4>Screenshots</h4><pre>Screenshots provided for all **required device sizes**:</pre><p>- iPhone 6.9&quot; (required)</p><p>- iPhone 6.5&quot; (required)</p><p>- iPad 13&quot; (required if supporting iPad)</p><p>- iPad 12.9&quot; 2nd gen (required if supporting iPad)</p><pre>Screenshots show **actual app UI** — not marketing graphics alone</pre><pre>No device frames required, but if used, use official Apple frames</pre><pre>Screenshots are in the correct **orientation** (portrait/landscape) matching the app</pre><h4>App Preview Video (optional but recommended)</h4><pre>15–30 seconds</pre><pre>Captured on device (or simulator), not animations</pre><pre>Audio is optional; if present, it should be appropriate</pre><h4>Description</h4><pre>**First 3 lines** (before &quot;more&quot;) are the strongest hook — this is what users see without tapping</pre><pre>Accurate description of features — do not mention features that are not in this version</pre><pre>No references to competitor apps or pricing of competitors</pre><pre>No mention of &quot;beta&quot;, &quot;free for limited time&quot; unless actively true</pre><pre>No grammatical errors</pre><h4>Keywords</h4><pre>Up to **100 characters** used strategically</pre><pre>No repetition of words already in the app name</pre><pre>No competitor names</pre><pre>Comma-separated, no spaces after commas</pre><h4>Subtitle</h4><pre>Max **30 characters**</pre><pre>Complements the app name without repeating it</pre><h4>Support URL</h4><pre>Points to a **real, working page** with actual support info</pre><pre>Not a generic homepage — should specifically address app support</pre><h4>Privacy Policy URL</h4><pre>**Mandatory** if the app collects any user data, or has Sign in with Apple, or is in certain categories</pre><pre>URL must be live and load correctly</pre><pre>Content must accurately describe what data is collected and how it is used</pre><p><strong><em>Primary and secondary category</em></strong><em> selected correctly — a mismatched category is a rejection reason.<br></em><strong><em>Copyright</em></strong><em> field filled in — format: © 2026 Your Name or Company</em></p><h4><strong><em>Export Compliance (Encryption)</em></strong></h4><p>ITSAppUsesNonExemptEncryption key added to Info.plist — set to NO if your app only uses standard HTTPS (most apps). Without this key, Apple will ask you manually on every build upload.<br>If your app uses custom encryption beyond HTTPS, set to YES and provide export compliance documentation.</p><blockquote>8. Privacy &amp; Data Collection</blockquote><pre>**App Privacy** section in App Store Connect filled out completely</pre><pre>Every data type collected is listed (email, device ID, usage data, etc.)</pre><pre>Data linked to user identity is correctly identified</pre><pre>If using third-party SDKs (Firebase, RevenueCat, Amplitude…), their data collection is included in your privacy labels</pre><pre>**ATT (App Tracking Transparency)** prompt implemented if you use IDFA or cross-app tracking (`NSUserTrackingUsageDescription` in `Info.plist`)</pre><pre>`PrivacyInfo.xcprivacy` file included if using any Apple-required-reason APIs (file timestamps, disk space, etc.)</pre><blockquote>9. In-App Purchases (if applicable)</blockquote><pre>All products created in App Store Connect with correct **Product IDs** matching the code</pre><pre>Products are in **&quot;Ready to Submit&quot;** status</pre><pre>A **screenshot of the IAP** uploaded in App Store Connect (required for review)</pre><pre>**Restore Purchases** button implemented and working</pre><pre>Free trial or introductory pricing configured correctly in both RevenueCat/StoreKit and App Store Connect</pre><pre>Paywall does **not** mislead the user about what is free vs. paid</pre><pre>`StoreKit` entitlement validation logic tested with a Sandbox account</pre><pre>Sandbox test account created in App Store Connect Users &amp; Roles</pre><blockquote>10. Sign in with Apple (if applicable)</blockquote><pre>**Sign in with Apple** capability added in Xcode</pre><pre>Enabled in the App ID on the Developer Portal</pre><pre>If any other social login (Google, Facebook, etc.) is offered, **Sign in with Apple must also be offered**</pre><pre>First/last name and email are only requested if actually used</pre><pre>Hiding email relay works and emails reach the user</pre><blockquote>11. App Review Guidelines Compliance</blockquote><pre>Read the [App Store Review Guidelines](https://developer.apple.com/app-store/review/guidelines/) — especially sections relevant to your app category</pre><pre>App does what the description says — **no hidden features**</pre><pre>No placeholder content, lorem ipsum, or broken screens</pre><pre>Login-required apps: **demo account credentials** provided in the Review Notes</pre><pre>Hardware-dependent features: explained in Review Notes if reviewer cannot test them (e.g. NFC, specific sensors)</pre><pre>No private API usage (use `nm`, `otool`, or Xcode&#39;s static analyzer to verify)</pre><pre>No undocumented URL scheme usage</pre><pre>Web content in `WKWebView` complies with guidelines — no full browser implementation</pre><pre>Kids category: strictly follows [kids app guidelines](https://developer.apple.com/app-store/review/guidelines/#kids) — no third-party analytics, no external links</pre><blockquote>12. Age Rating</blockquote><pre>Age rating questionnaire completed accurately in App Store Connect</pre><pre>If app contains user-generated content: **UGC moderation** mechanism in place</pre><pre>Mature content (violence, adult themes) declared honestly</pre><blockquote>13. TestFlight — Internal Testing</blockquote><pre>Build uploaded and processed without errors</pre><pre>Tested on a **real device** (not only simulator) on the minimum deployment target</pre><pre>Tested on the **latest iOS version** as well</pre><pre>All critical flows tested end-to-end:</pre><p>- [ ] Onboarding / first launch</p><p>- [ ] Core feature loop</p><p>- [ ] IAP purchase and restore</p><p>- [ ] Sign in / Sign out</p><p>- [ ] Push notification opt-in</p><p>- [ ] Offline behavior</p><pre>App does not crash on launch</pre><pre>No missing assets (broken images, missing fonts)</pre><pre>All external links open correctly</pre><pre>Deep links / Universal Links work</pre><blockquote>14. TestFlight — External Testing (optional but recommended)</blockquote><pre>Beta App Review approved for external TestFlight</pre><pre>At least 5–10 external testers have validated the core flows</pre><pre>Feedback from testers addressed before submission</pre><blockquote>15. Final Build &amp; Upload</blockquote><pre>Built with the **Release** scheme (not Debug)</pre><pre>Bitcode setting matches App Store Connect expectations (disabled for Xcode 14+)</pre><pre>Archived via `Product → Archive` in Xcode</pre><pre>Validated against App Store in Organizer — **no errors, no warnings**</pre><pre><em>Uploaded to App Store Connect via Organizer or xcodebuild -exportArchive / Transporter app</em></pre><pre>Build appears in App Store Connect under the correct version</pre><blockquote>16. App Store Connect — Version Submission</blockquote><pre>Correct build selected for the version</pre><pre>**&quot;What&#39;s New&quot;** text written (for updates; not required for 1.0 but good practice)</pre><pre>Release type selected:</pre><p>- Manually release this version — you control when it goes live after approval</p><p>- Automatically release this version — goes live immediately after approval (default)</p><p>- Scheduled release — set a date</p><pre>**Phased release** considered for updates (rolls out to 1% → 100% over 7 days)</pre><pre>Review Notes filled in if the app has any non-obvious features or requires a test account</pre><p><strong>Content Rights</strong> declaration completed — if your app displays third-party content, you must confirm you have the rights (toggle in App Store Connect on the version page)</p><blockquote>17. Submit for Review</blockquote><pre>Everything above is checked</pre><pre>`Submit for Review` clicked</pre><pre>Confirmation email received from Apple</pre><blockquote>18. Post-Submission</blockquote><pre>Monitor App Store Connect for status changes (In Review → Approved / Rejected)</pre><pre>If rejected: read the rejection reason carefully, respond via the Resolution Center — do not resubmit without addressing every point</pre><pre>Once approved and released: verify the app is live and the App Store listing looks correct</pre><pre>Set up **crash monitoring** (Xcode Organizer, Firebase Crashlytics, Sentry) and check within 24h of launch</pre><pre>Monitor **reviews** and respond promptly</pre><blockquote>Quick Reference: Most Common Rejection Reasons</blockquote><p>Reason | Fix</p><p>Guideline 2.1 — Crashes / Bugs | Test more thoroughly on real devices</p><p>Guideline 3.1.1 — IAP not using Apple&#39;s system | All digital goods must use StoreKit</p><p>Guideline 5.1.1 — Privacy policy missing | Add a real privacy policy URL</p><p>Guideline 4.0 — Copycat / low quality | Add meaningful functionality</p><p>Guideline 1.2 — Developer info incomplete | Fill all contact fields in App Store Connect</p><p>Guideline 2.3 — Inaccurate metadata | Match screenshots and description to actual app</p><p>Permission usage description missing | Add NSXxxUsageDescription to Info.plist</p><p>Sign in with Apple missing | Required when any third-party login is present</p><p>Ship confidently. Every item on this list exists because someone got rejected for it.</p><h4>Once your app is live on the App Store, don’t just celebrate — promote it. Head over to <a href="https://appfra.me/">https://appfra.me</a> to generate eye-catching social media visuals for your app in seconds.</h4><p>Good luck shipping!.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d16d1645e7da" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Claude Code Has a Hidden Setting That Makes It 600x Faster]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@simonsruggi/claude-code-has-a-hidden-setting-that-makes-it-600x-faster-eaaad67a25a1?source=rss-880181e9f4ec------2"><img src="https://cdn-images-1.medium.com/max/1200/1*bLV2B_JVaHorm6uLU0xTWw.png" width="1200"></a></p><p class="medium-feed-snippet">Most people using Claude Code don&apos;t know this setting exists. And it&apos;s the single biggest performance upgrade you can make.</p><p class="medium-feed-link"><a href="https://medium.com/@simonsruggi/claude-code-has-a-hidden-setting-that-makes-it-600x-faster-eaaad67a25a1?source=rss-880181e9f4ec------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@simonsruggi/claude-code-has-a-hidden-setting-that-makes-it-600x-faster-eaaad67a25a1?source=rss-880181e9f4ec------2</link>
            <guid isPermaLink="false">https://medium.com/p/eaaad67a25a1</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[claude-code]]></category>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[productivity]]></category>
            <dc:creator><![CDATA[Simone Ruggiero]]></dc:creator>
            <pubDate>Sun, 08 Mar 2026 06:43:56 GMT</pubDate>
            <atom:updated>2026-03-08T06:43:56.230Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[The Truth Behind Creating an iOS App in One Week With AI]]></title>
            <link>https://medium.com/@simonsruggi/the-truth-behind-creating-an-ios-app-in-one-week-with-ai-9e0940d5314a?source=rss-880181e9f4ec------2</link>
            <guid isPermaLink="false">https://medium.com/p/9e0940d5314a</guid>
            <category><![CDATA[productivity]]></category>
            <category><![CDATA[ios-development]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[swift]]></category>
            <dc:creator><![CDATA[Simone Ruggiero]]></dc:creator>
            <pubDate>Sat, 07 Mar 2026 21:42:02 GMT</pubDate>
            <atom:updated>2026-03-07T21:42:02.378Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Q-MdMUKtfPK1zJ-l8e5GSw.png" /><figcaption>ZenFlip</figcaption></figure><p>I built a full iOS app with Claude Code — here&#39;s what actually happened.</p><p>It’s free on the App Store: <a href="https://apps.apple.com/app/zenflip/id6758057185">HERE</a></p><p>Not a toy project. Not a tutorial demo. A real app, on the App Store, with 6,000+ lines of Swift, CoreMotion sensor integration, Dynamic Island support, Game Center leaderboards, RevenueCat subscriptions, and a proper MVVM architecture.</p><p>The app is called ZenFlip — and if you struggle with phone addiction or staying focused, it might be for you too.</p><h4>The Problem ZenFlip Solves</h4><p>We all know the cycle. You sit down to work, grab your phone &quot;just to check one thing,&quot; and 40 minutes later you&#39;re deep into Instagram.</p><p>The problem isn&#39;t willpower — it&#39;s that your phone is right there, screen up, notifications glowing, begging you to pick it up.</p><p>ZenFlip turns that problem into a solution. <strong>Flip your phone face-down, and a focus timer starts automatically.</strong> Pick it up, it pauses. No buttons to press, no app to open, no setup. Just flip and focus.</p><p>The physical gesture matters. It&#39;s a commitment — you&#39;re literally turning away from distractions. And because the app tracks your streaks (consecutive days of hitting your focus goal), it builds a habit loop. Miss a day, lose your streak. It&#39;s simple psychology, and it works.</p><p>Here&#39;s what you get:</p><p>- <strong>Flip detection</strong> — place your phone face-down and the timer starts. Pick it up, it pauses. That&#39;s it.</p><p>- <strong>Dynamic Island</strong> — your active timer shows right on the lock screen and Dynamic Island, so you always know how long you&#39;ve been focused without unlocking your phone.</p><p>- <strong>Daily streaks</strong> — set a goal (15 minutes to 2 hours) and maintain your streak by hitting it every day.</p><p>- <strong>Weekly stats</strong> — see your focus patterns, total time, and session history.</p><p>- <strong>Leaderboard</strong> — compete with other users on focus time and longest streaks via Game Center.</p><p>- <strong>Pomodoro mode</strong> (Premium) — classic 25/5 work-break cycles with automatic transitions.</p><p>- <strong>Custom themes</strong> (Premium) — Dark, Light, Purple, Ocean, Forest.</p><p>You can download ZenFlip for free on the App Store. The core experience — flip, focus, streaks — is completely free. Premium unlocks Pomodoro mode and themes.</p><p>Now, here&#39;s how Claude Code helped me build the whole thing.</p><h4>What&#39;s Under the Hood</h4><p>The concept sounds simple, but &quot;flip your phone to start a timer&quot; has real engineering challenges.</p><p>Your phone&#39;s accelerometer reports a gravity vector across three axes. When gravity.z approaches 1.0, the screen is facing down. But you can&#39;t just check that value once — the real app needed:</p><p>- <strong>Anti-cheat detection</strong> — people will tilt their phone at weird angles to game the streak system. The motion service uses strict thresholds (0.92 for face-down) plus tilt tolerance checks on x/y axes, plus stability variance monitoring. You can&#39;t just wave your phone around.</p><p>- <strong>Debounce logic</strong> — 10 consecutive readings at 20Hz (0.5 seconds) before starting a session. 6 readings (0.3 seconds) to stop. Fast enough to feel responsive, slow enough to ignore accidental bumps.</p><p>- <strong>Dynamic Island integration</strong> — Live Activities showing your timer right on the lock screen and Dynamic Island. This alone required its own ActivityAttributes model and careful state management.</p><p>- <strong>Streak system with midnight resets</strong> — tracking consecutive days, handling timezone edge cases, persisting across app restarts.</p><p>- <strong>Leaderboards via Game Center</strong> — competing on total focus time and longest streaks.</p><p>- <strong>RevenueCat subscriptions</strong> — Pomodoro mode, custom themes, and advanced goals behind a paywall.</p><p>That&#39;s 25+ Swift files across Models, ViewModels, Views, Services, and Utilities.</p><h4>How Claude Code Actually Helped</h4><p>I didn&#39;t just say &quot;build me an app.&quot; That&#39;s not how it works. Here&#39;s what actually happened.</p><h4>1. Architecture Decisions</h4><p>I described the app concept and Claude Code suggested the MVVM structure immediately — separating MotionService, TimerViewModel, and the SwiftUI views. It set up the file structure, the Combine publishers for reactive state, and the dependency injection pattern.</p><p>Was it perfect from the start? No. But the bones were right, and iterating on a solid foundation is 10x faster than refactoring a mess.</p><h4>2. The Motion Detection Algorithm</h4><p>This is where Claude Code genuinely impressed me. The accelerometer logic isn&#39;t trivial — you need to handle:</p><p>- Gravity vector decomposition across three axes</p><p>- Consecutive reading thresholds for debouncing</p><p>- Stability variance to detect if the phone is actually still</p><p>- Different sensitivity for starting vs stopping (anti-cheat vs responsiveness)</p><p>Claude Code wrote the initial MotionService with all of these considerations. I tested on device, found edge cases (phone on a soft couch registers micro-movements), gave feedback, and Claude Code adjusted the thresholds. Three iterations and it was solid.</p><h4>3. Dynamic Island &amp; Live Activities</h4><p>This was the hardest part. Live Activities on iOS have very specific requirements — ActivityAttributes, the update/end lifecycle, the compact vs expanded layouts. The documentation is dense and the API changed across iOS versions.</p><p>Claude Code handled the boilerplate, the attribute definitions, and the state synchronization between the timer and the Live Activity. I focused on making it look right.</p><h4>4. The Stuff That Would Have Taken Forever</h4><p>- Setting up RevenueCat integration with StoreKit 2</p><p>- Building the paywall view with subscription options</p><p>- Game Center authentication and leaderboard submission</p><p>- Statistics calculations with weekly aggregation</p><p>- Sound service for session start/end notifications</p><p>- Review request manager with smart timing</p><p>- Onboarding flow</p><p>- Settings with customizable daily goals</p><p>Each of these is maybe 1-2 hours of work individually. Together, they&#39;re a week of boilerplate. Claude Code knocked them out while I focused on the UX and the parts that actually needed human judgment.</p><h4>Where Claude Code Fell Short</h4><p>I&#39;d be lying if I said it was flawless.</p><p>- <strong>UI polish</strong> — Claude Code can build functional SwiftUI views, but the fine-tuning of animations, spacing, and &quot;feel&quot; required my eye and manual iteration. AI doesn&#39;t have taste yet.</p><p>- <strong>App Store edge cases</strong> — provisioning profiles, entitlements configuration, specific App Store Connect settings. Claude Code knows the concepts but can&#39;t click buttons in Xcode for you.</p><p>- <strong>On-device testing</strong> — the motion detection needed real-world testing on actual hardware. No simulator can replicate someone putting their phone on a wobbly desk.</p><h4>The Real Takeaway</h4><p>Claude Code didn&#39;t replace me. It amplified me.</p><p>The parts I&#39;m good at — UX decisions, visual polish, understanding what users actually want — I still did those. The parts that are just <em>work</em> — wiring up services, implementing known patterns, writing boilerplate — Claude Code handled.</p><p>ZenFlip has 6,000 lines of Swift. I probably wrote 2,000 of them by hand. The other 4,000? Claude Code wrote them, I reviewed them, I tested them, and I iterated on them.</p><p>The result is an app that would have taken me 3-4 weeks working alone. It took one week.</p><p>That&#39;s not cheating. That&#39;s leverage.</p><h4>Try ZenFlip</h4><p>If you&#39;re tired of losing hours to your phone, give ZenFlip a try. Flip your phone, focus on what matters, build your streak.</p><p>It&#39;s free on the App Store: <a href="https://apps.apple.com/app/zenflip/id6758057185">HERE</a></p><p>The best productivity system is the one you&#39;ll actually use. And this one only asks you to do one thing: put your phone down.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9e0940d5314a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[I Built a Free, Open-Source macOS Menu Bar App to Track Stocks — Here’s Why]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@simonsruggi/i-built-a-free-open-source-macos-menu-bar-app-to-track-stocks-heres-why-07cbc0d3e631?source=rss-880181e9f4ec------2"><img src="https://cdn-images-1.medium.com/max/2600/1*wMZ_4WxY28kJ0WYNQnQUOw.png" width="3004"></a></p><p class="medium-feed-snippet">If you&#x2019;re like me, you check your stock portfolio way too many times a day. I used to open Yahoo Finance, switch between browser tabs, or&#x2026;</p><p class="medium-feed-link"><a href="https://medium.com/@simonsruggi/i-built-a-free-open-source-macos-menu-bar-app-to-track-stocks-heres-why-07cbc0d3e631?source=rss-880181e9f4ec------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@simonsruggi/i-built-a-free-open-source-macos-menu-bar-app-to-track-stocks-heres-why-07cbc0d3e631?source=rss-880181e9f4ec------2</link>
            <guid isPermaLink="false">https://medium.com/p/07cbc0d3e631</guid>
            <category><![CDATA[swiftui]]></category>
            <category><![CDATA[mac-os-x]]></category>
            <category><![CDATA[open-source]]></category>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[stock-market]]></category>
            <dc:creator><![CDATA[Simone Ruggiero]]></dc:creator>
            <pubDate>Wed, 04 Mar 2026 19:07:37 GMT</pubDate>
            <atom:updated>2026-03-04T19:07:37.802Z</atom:updated>
        </item>
    </channel>
</rss>