<?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 AI Book Shelf on Medium]]></title>
        <description><![CDATA[Stories by AI Book Shelf on Medium]]></description>
        <link>https://medium.com/@ai_book_shelf?source=rss-72ccc1abbc02------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*dmbNkD5D-u45r44go_cf0g.png</url>
            <title>Stories by AI Book Shelf on Medium</title>
            <link>https://medium.com/@ai_book_shelf?source=rss-72ccc1abbc02------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 00:06:05 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@ai_book_shelf/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[How to start an emergency fund]]></title>
            <link>https://medium.com/@ai_book_shelf/how-to-start-an-emergency-fund-63862a871c50?source=rss-72ccc1abbc02------2</link>
            <guid isPermaLink="false">https://medium.com/p/63862a871c50</guid>
            <category><![CDATA[budgeting]]></category>
            <category><![CDATA[saving-money]]></category>
            <category><![CDATA[personal-finance]]></category>
            <category><![CDATA[financial-planning]]></category>
            <category><![CDATA[money]]></category>
            <dc:creator><![CDATA[AI Book Shelf]]></dc:creator>
            <pubDate>Thu, 19 Nov 2026 06:04:24 GMT</pubDate>
            <atom:updated>2026-11-19T06:04:24.263Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/800/0*oYj7TZRKXBymG0Vf" /></figure><p>An emergency fund is easier to start than the internet makes it sound, and easier to abandon than it looks on a spreadsheet. The usual advice — three to six months of expenses — is a destination, not a first step. If that number feels impossible, you will not open the account.</p><p>This is general education, not financial, tax, or legal advice. Your situation may include debt, irregular income, or benefits that change the picture. Use this as a starting method, then check it against your own numbers.</p><h3>Pick a first number you will actually hit</h3><p>A useful first target is one modest, named amount: $500, $1,000, or one month of rent plus a utility bill, whichever you can reach in a few months without skipping groceries. Write the number down. A target you can finish teaches the habit of keeping a buffer. A target you cannot finish stays theoretical.</p><p>If your income jumps around, set the first target from a low month, not an optimistic average. The fund is there for the low months.</p><h3>Give the money a separate place to sit</h3><p>The starter buffer should not live in the same checking account you spend from. A separate savings account at the same bank is enough for a first fund. The goal is a small amount of friction, not a clever investment.</p><p>Do not put the starter buffer in the stock market or in anything you would have to sell in a hurry. This money’s job is to be boring and available. Yield can wait until the fund is large enough that a day or two of transfer time is acceptable.</p><h3>Find the first few hundred, then automate</h3><p>Look once, on paper, at last month’s actual spending. You are not hunting for moral failures. You are looking for one or two transfers that can move automatically the day after payday: a leftover subscription, a repeating transfer that used to go to a vague “savings” pot, or a small amount that still leaves the month livable.</p><p>Automate that transfer. If a month is tight, reduce the amount; do not turn the automation off and promise to restart later. A smaller automatic transfer beats a heroic one that happens twice.</p><h3>Decide in advance what counts as an emergency</h3><p>Write a short list: a job loss, an urgent medical bill, a required car repair that would otherwise go on a credit card, a security deposit you must pay to not become unhoused. “I am tired and want a weekend away” is not on the list. Neither is a sale.</p><p>When you do spend from the fund, record what it was for and start the refill on the next payday, even at a lower rate. A fund you empty and then ignore was a one-time stash, not a buffer.</p><p>If you also have high-interest debt, a tiny starter buffer still comes first for most people, so the next surprise does not go back on the card. After that, the split between debt payments and growing the fund is a personal judgment, not a slogan.</p><h3>A book that walks through the first fund and the later one</h3><p><a href="https://ai-book-shelf-library.whop.site/books/emergency-fund-first.html">Emergency Fund First</a> covers picking the first number, where the money sits, automation that does not starve the month, what counts as an emergency, and how to graduate from a starter buffer to a fuller fund. It is general education, not a plan for your household.</p><p>You can <a href="https://ai-book-shelf-library.whop.site/read/emergency-fund-first.html">read the opening chapters free</a>. The complete book is $3.99, one-time, or you can <a href="https://ai-book-shelf-library.whop.site/membership.html">read every book for $9.99/month</a>.</p><p>From the shop</p><h3>Emergency Fund First</h3><p>Build a starter buffer you will actually finish.</p><p>Or <a href="https://ai-book-shelf-library.whop.site/membership.html">read every book for $9.99/month</a>.</p><p><em>Originally published at https://ai-book-shelf-library.whop.site.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=63862a871c50" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[A script for asking for a raise]]></title>
            <link>https://medium.com/@ai_book_shelf/a-script-for-asking-for-a-raise-2e16e93b60de?source=rss-72ccc1abbc02------2</link>
            <guid isPermaLink="false">https://medium.com/p/2e16e93b60de</guid>
            <category><![CDATA[career-advice]]></category>
            <category><![CDATA[personal-finance]]></category>
            <category><![CDATA[workplace]]></category>
            <category><![CDATA[communication-skills]]></category>
            <category><![CDATA[salary-negotiations]]></category>
            <dc:creator><![CDATA[AI Book Shelf]]></dc:creator>
            <pubDate>Wed, 18 Nov 2026 06:02:21 GMT</pubDate>
            <atom:updated>2026-11-18T06:02:21.649Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/800/0*VK2AjYMCzprJREP4" /></figure><p>People often wait to ask for a raise until they feel brave enough. Bravery is unreliable. What helps is a small packet of numbers, a time that makes sense for the company, and a script you have said out loud once before the meeting.</p><p>This is general career guidance, not legal, tax, or financial advice, and it will not fit every workplace. If your company has a formal compensation cycle, use that calendar. If you are in a union or on a written contract, follow those rules first.</p><h3>Do the math before you ask for the meeting</h3><p>A raise request is easier to hear when it is about contribution and market, not about a rent increase. Gather two kinds of numbers:</p><ul><li>Contribution: what you own, what changed because of your work, and a denominator (time saved, errors reduced, revenue influenced, projects shipped). Rough and honest beats precise and inflated.</li><li>Market: two or three current references for the same level and region, from job posts, a recruiter range, or a salary band your company already publishes. Note the date and the source.</li></ul><p>Put this on one page. If it will not fit on one page, you are still collecting, not ready to ask. The page is for you first. You may share a shorter version in the meeting.</p><h3>Ask for a meeting, not a decision in the hallway</h3><p>Do not open with the number in Slack. Ask for time: “I would like twenty minutes to talk about my role and compensation. I will send a one-page summary the day before.” That sentence is the first script. It gives your manager a chance to arrive prepared, which is in your interest.</p><p>Timing matters. A week after you saved a difficult launch is better than the morning a budget freeze is announced, and better than the Friday before your manager’s vacation. If a formal review is six weeks away, ask whether this conversation should feed that review rather than sit outside it.</p><h3>A script you can say without sounding stiff</h3><p>You do not need to memorize a speech. You need four beats you can hit in order:</p><ul><li>Scope: “Over the past year my work has included [two or three owned outcomes].”</li><li>Evidence: “The change I can point to is [one number with a denominator], and [one qualitative result a peer would recognize].”</li><li>Ask: “I am requesting an adjustment to [amount or band], which sits inside the range I have seen for this level.”</li><li>Pause: stop talking. Let your manager respond. If you fill the silence, you start bargaining against yourself.</li></ul><p>If you are offered praise and no number, ask a calm follow-up: “What would need to be true for this to be approved, and by when can we revisit it?” Write down whatever they say.</p><h3>When the answer is no, delayed, or partial</h3><p>A no is information. Ask whether it is a budget no, a level no, or a timing no. Those are different. A budget no might still allow a title, a bonus, or a dated review. A level no should come with the missing evidence, written down, not implied.</p><p>Do not accept a vague “later” without a date and a criterion. A dated follow-up is a professional request, not a confrontation.</p><h3>A book with the packet, the scripts, and worked numbers</h3><p><a href="https://ai-book-shelf-library.whop.site/books/raise-request-math.html">Raise Request Math</a> is a short book on this process: market anchors, a one-page packet with worked numbers, rehearsal, pushbacks, and what to do after a partial yes. It is general education, not a promise that any workplace will say yes.</p><p>You can <a href="https://ai-book-shelf-library.whop.site/read/raise-request-math.html">read the opening chapters free</a>. The complete book is $3.99, one-time, or you can <a href="https://ai-book-shelf-library.whop.site/membership.html">read every book for $9.99/month</a>.</p><p>From the shop</p><h3>Raise Request Math</h3><p>Prepare the numbers before you say raise.</p><p>Or <a href="https://ai-book-shelf-library.whop.site/membership.html">read every book for $9.99/month</a>.</p><p><em>Originally published at </em><a href="https://ai-book-shelf-library.whop.site/articles/script-for-asking-for-a-raise.html"><em>https://ai-book-shelf-library.whop.site</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2e16e93b60de" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to cut back-to-back meetings]]></title>
            <link>https://medium.com/@ai_book_shelf/how-to-cut-back-to-back-meetings-e3ac55364788?source=rss-72ccc1abbc02------2</link>
            <guid isPermaLink="false">https://medium.com/p/e3ac55364788</guid>
            <category><![CDATA[self-improvement]]></category>
            <category><![CDATA[workplace]]></category>
            <category><![CDATA[meetings]]></category>
            <category><![CDATA[time-management]]></category>
            <category><![CDATA[productivity]]></category>
            <dc:creator><![CDATA[AI Book Shelf]]></dc:creator>
            <pubDate>Tue, 17 Nov 2026 05:57:50 GMT</pubDate>
            <atom:updated>2026-11-17T05:57:50.883Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/800/0*5tFzErT0wCmOFKhY" /></figure><p>Back-to-back meetings feel like work, which is why they survive. A calendar that is full looks responsible. It is often just a stack of status circles, leftover recurring invites, and “got a sec?” conversations that should have been a written decision.</p><p>You will not delete your way to a quiet week in a single afternoon, especially if you are not the most senior person in the room. You can, though, run a filter, convert some meetings into documents, and use a short script when you decline.</p><h3>Inventory the recurring invites first</h3><p>List every recurring meeting you attend in a normal week, including the ones you join late and the ones you keep “just in case.” For each one, write the purpose in one sentence. If you cannot, that is already a signal.</p><p>Then put each invite through a keep / kill / convert filter:</p><ul><li>Keep: a decision must be made, or a relationship genuinely requires live time (a one-to-one, a genuine working session, a meeting with a person who cannot write comfortably in your language).</li><li>Kill: the purpose is habit, the same update is already in a channel, or you are there only to be seen.</li><li>Convert: the purpose is status, review, or a decision that can be made on a document with a deadline and a named decision-maker.</li></ul><p>You do not have to win every conversion. Start with the meetings you own. For meetings you do not own, a useful first step is a question, not a cancellation: “Would a written update before Friday work instead of the half hour?”</p><h3>Replace status circles with a decision document</h3><p>A decision document can be short: the decision needed, the options, the recommendation, the date it will be considered final if no one objects, and who owns the call. People can comment when they have focus. The meeting, if it still happens, starts from the comments instead of from a blank circle.</p><p>Office hours can absorb the “got a sec?” pile. Two or three short windows a week, on the calendar, are easier on everyone than a day of interruptions. People with a real blocker still get live time. The rest get a written answer.</p><h3>Scripts for declining or converting</h3><p>A decline should be brief, specific, and helpful:</p><ul><li>“I will skip this one and read the notes. If a decision needs me, please flag me on the doc by Thursday.”</li><li>“I can do 15 minutes at the start for the decision, then drop off. I will send my update in writing beforehand.”</li><li>“This looks like a status pass. I can post a written update each Monday instead of attending.”</li></ul><p>You will sometimes be told to come anyway. Go, once, and then ask after the meeting whether the written version would have been enough. The goal is recovered hours, not a reputation for being difficult.</p><h3>Keep a modest hours log</h3><p>For two weeks, mark meetings you declined, converted, or shortened, and guess the time returned. You are not building a performance dashboard. You are checking whether the filter is doing anything. If the number does not move, you are still attending things you already classified as convert or kill.</p><h3>A book with the filter, the scripts, and a thirty-day diet</h3><p><a href="https://ai-book-shelf-library.whop.site/books/meeting-less.html">Meeting-Less</a> is a short book on this: the keep / kill / convert filter, decision docs, office hours, time zones, stakeholder meetings you cannot skip, and a thirty-day meeting diet. It will not make every workplace async overnight.</p><p>You can <a href="https://ai-book-shelf-library.whop.site/read/meeting-less.html">read the opening chapters free</a>. The complete book is $4.99, one-time, or you can <a href="https://ai-book-shelf-library.whop.site/membership.html">read every book for $9.99/month</a>.</p><p>From the shop</p><h3>Meeting-Less</h3><p>Reclaim hours with async-first collaboration.</p><p>Or <a href="https://ai-book-shelf-library.whop.site/membership.html">read every book for $9.99/month</a>.</p><p><em>Originally published at </em><a href="https://ai-book-shelf-library.whop.site/articles/how-to-cut-back-to-back-meetings.html"><em>https://ai-book-shelf-library.whop.site</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e3ac55364788" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to collect evidence for a promotion all year]]></title>
            <link>https://medium.com/@ai_book_shelf/how-to-collect-evidence-for-a-promotion-all-year-7c1e4d3af297?source=rss-72ccc1abbc02------2</link>
            <guid isPermaLink="false">https://medium.com/p/7c1e4d3af297</guid>
            <category><![CDATA[workplace]]></category>
            <category><![CDATA[promotion]]></category>
            <category><![CDATA[self-improvement]]></category>
            <category><![CDATA[careers]]></category>
            <category><![CDATA[productivity]]></category>
            <dc:creator><![CDATA[AI Book Shelf]]></dc:creator>
            <pubDate>Mon, 16 Nov 2026 05:52:49 GMT</pubDate>
            <atom:updated>2026-11-16T05:52:49.457Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/800/0*Ejri2FMxt1gR0bQp" /></figure><p>Promotion packets go badly when they are written from memory on a Sunday night. You remember the dramatic week and forget the work that was quiet, shared, or finished in March. Managers forget too. A thin packet is often a record-keeping problem, not a talent problem.</p><p>The fix is a folder you add to in five-minute bursts, close to the work, all year — or from this week forward, even if the year is already half gone.</p><h3>Decide what counts as evidence</h3><p>Useful evidence is specific and checkable. A compliment in passing is nice and usually not enough. Prefer:</p><ul><li>Outcomes with a denominator: “cut the weekly close from two days to four hours,” not “improved the close.”</li><li>Scope and ownership: what you were responsible for, where it started, and where it ended. Shared work should name the others and still say what you owned.</li><li>Artifacts: a design, a decision record, a before-and-after metric, a customer note, a launch plan. Keep the thing, not only your memory of the thing.</li><li>Invisible work that kept something from breaking, with enough detail that a reader who was not in the room can see it.</li></ul><p>If you cannot point to a date, a result, or a document, it is still a story. Stories can support a packet. They should not be the whole packet.</p><h3>Make a folder you will actually keep</h3><p>One folder, in the tools you already use — a drive directory, a notes page, even a running document — with a simple shape: a file or heading per quarter, and a short log at the top. If the system needs a Saturday to maintain, you will not maintain it.</p><p>When you finish something, spend five minutes: what happened, what changed, who else was involved, and a link to the artifact. Do it the same day if you can. Evidence decays faster than it feels like it does.</p><h3>Review once a quarter so the folder stays honest</h3><p>Every quarter, spend half an hour rereading your own notes. Drop the items that were activity without a result. Strengthen the ones that now have a better number. Notice gaps: if your job requires mentoring or cross-team work and the folder has none, that is a signal to collect it going forward, not to invent it.</p><p>Confidential work can be described at the right altitude: the problem class, the result, and who can vouch, without pasting a private document into a packet.</p><h3>Write the packet from the folder, not from a blank page</h3><p>When the form arrives, you are sorting and tightening, not inventing a year. Start with the strongest three to five pieces of evidence that match the level you are aiming at. Map each one to the company’s language if it has a rubric. Leave out the rest, or move it to a backup list your manager can use in discussion.</p><p>If you are starting in November, do not fake an archive. Collect from here, recover what you can from sent mail and shipped artifacts, and be honest about the date range. A clear six months is more useful than a fuzzy year.</p><h3>A book with the folder method and the hard cases</h3><p><a href="https://ai-book-shelf-library.whop.site/books/promo-evidence-folder.html">Promo Evidence Folder</a> covers what counts as evidence, a folder you will keep, outcomes with a denominator, invisible and shared work, a quarterly review, and how to start mid-year without inventing an archive.</p><p>You can <a href="https://ai-book-shelf-library.whop.site/read/promo-evidence-folder.html">read the opening chapters free</a>. The complete book is $4.99, one-time, or you can <a href="https://ai-book-shelf-library.whop.site/membership.html">read every book for $9.99/month</a>.</p><p>From the shop</p><h3>Promo Evidence Folder</h3><p>Collect proof all year so promotion packets write themselves.</p><p>Or <a href="https://ai-book-shelf-library.whop.site/membership.html">read every book for $9.99/month</a>.</p><p><em>Originally published at </em><a href="https://ai-book-shelf-library.whop.site/articles/how-to-collect-evidence-for-a-promotion.html"><em>https://ai-book-shelf-library.whop.site</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7c1e4d3af297" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to end the workday so the evening actually starts]]></title>
            <link>https://medium.com/@ai_book_shelf/how-to-end-the-workday-so-the-evening-actually-starts-1fd3cb6dbb68?source=rss-72ccc1abbc02------2</link>
            <guid isPermaLink="false">https://medium.com/p/1fd3cb6dbb68</guid>
            <category><![CDATA[productivity]]></category>
            <category><![CDATA[remote-work]]></category>
            <category><![CDATA[work-life-balance]]></category>
            <category><![CDATA[habits]]></category>
            <category><![CDATA[mental-health]]></category>
            <dc:creator><![CDATA[AI Book Shelf]]></dc:creator>
            <pubDate>Sun, 15 Nov 2026 09:03:39 GMT</pubDate>
            <atom:updated>2026-11-15T09:03:39.232Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/800/0*HgSSlmi2dLmYyVSd" /></figure><p>A lot of evenings never really begin. The laptop lid closes, but the last unfinished item follows you into the kitchen. You check one more message while the food heats up. You tell yourself you will start the evening in ten minutes, and then it is time to sleep.</p><p>Ending the workday is a skill, not a mood. You do not wait until you feel finished. You install a small close-down that you can repeat on ordinary Tuesdays, including the days that ran long.</p><h3>Pick a stop time you can defend</h3><p>Choose a time that is slightly earlier than the hour you usually drift past. If you typically fade out at 6:40, try 6:15. The point is not to become a person who never works late. The point is to have a default you can name when someone asks for “just one more thing.”</p><p>Write the stop time where you will see it at 4 p.m., not only in a Saturday journal. A calendar alert titled “begin shutdown” works better than a vague intention to wrap up.</p><p>When the alert fires, you start the close-down even if a task is mid-sentence. That is the whole idea. Unfinished work gets parked, not secretly continued in your head.</p><h3>Park the work instead of carrying it</h3><p>Keep a parking list — a single note, not a second to-do app — and use it only at shutdown. For each open item, write one line:</p><ul><li>What it is, in plain language (not a filename).</li><li>The next physical action, small enough to start tomorrow without rereading a thread.</li><li>Whether it is waiting on someone else.</li></ul><p>This is not a complete project plan. It is a way to tell your brain the item has a place to live overnight. If you cannot name the next action, write “open the doc and find the decision” — that is still a next action.</p><h3>Take five minutes to hand the day to tomorrow</h3><p>After the parking list, spend five minutes on a short handoff to your future self. Tomorrow-you should not have to reconstruct today from a pile of tabs.</p><ul><li>Circle the first thing you will start tomorrow, not the most impressive thing.</li><li>Leave one sentence about anything a coworker might need if they look for you after you log off.</li><li>Close or mute the channels you do not need overnight, without announcing a dramatic disappearance.</li></ul><p>If you manage other people, the handoff includes them. A brief status in the shared place they already look is enough. A late-evening chat that “just confirms” usually reopens the day.</p><h3>Protect the first two hours after you stop</h3><p>The shutdown fails most often in the first two hours. That is when “one more email” feels harmless. Decide in advance what those hours are for: dinner, a walk, a book, talking to the people you live with. The activity matters less than not sliding back into the parking list.</p><p>When a genuine emergency arrives after the stop time, handle it, then do a twenty-minute mini-shutdown: update the parking list, name tomorrow’s first step, and close the channels again. A broken day is not a reason to abandon the ritual until Monday.</p><h3>A book that walks through the ritual</h3><p>This page is a starting method, not a full system. <a href="https://ai-book-shelf-library.whop.site/books/shutdown-complete-ritual.html">Shutdown Complete Ritual</a> goes further: how to defend a stop time with coworkers, what to do when other people wait on you, how to handle shared space at home, and a fifteen-weekday stretch for installing the habit.</p><p>You can <a href="https://ai-book-shelf-library.whop.site/read/shutdown-complete-ritual.html">read the opening chapters free</a>, with no account. The complete book is $4.99, one-time, or you can <a href="https://ai-book-shelf-library.whop.site/membership.html">read every book for $9.99/month</a>.</p><p>From the shop</p><h3>Shutdown Complete Ritual</h3><p>End the workday so evenings actually restore.</p><p>Or <a href="https://ai-book-shelf-library.whop.site/membership.html">read every book for $9.99/month</a>.</p><p><em>Originally published at </em><a href="https://ai-book-shelf-library.whop.site/articles/how-to-end-the-workday.html"><em>https://ai-book-shelf-library.whop.site</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1fd3cb6dbb68" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>