Hey, it’s Jeffrey — back again!
Tax-loss harvesting is one of those phrases I've heard about but never actually done. To be honest, I wouldn’t consider myself an expert in this area. The reason for that: most of my investments are in tax-advantaged retirement funds where tax-loss harvesting doesn’t work
Recently, I started investing more in a taxable brokerage account. So I finally thought: do I actually have an opportunity to do this now?
So I built a prompt to check. The idea is simple. Paste your raw brokerage export, answer a few basic questions about your filing status and income, and let AI do the math to see if you have a real opportunity.
Here's the prompt I used:
I want to check whether I currently have any tax-loss harvesting opportunity in my taxable brokerage account (not retirement accounts, this only applies to taxable).
I'm going to paste my raw position or lot data below, exactly as exported or copied from my brokerage. It may be messy or in an unfamiliar format. Figure out the relevant columns yourself (symbol, quantity, cost basis, current value, purchase date if available) rather than asking me to reformat it.
[PASTE RAW CSV OR COPIED TABLE HERE]
My filing status: [single / married filing jointly / married filing separately / head of household]
If single, head of household, or married filing separately: my own approximate income this year: $[X]
If married filing jointly: our approximate household income this year: $[X]
My state: [state]
Have I sold any investments for a gain this year: [yes, roughly $X / no / not sure]
Using this data:
Identify every position or lot currently sitting at an unrealized loss. Ignore cash and money market positions.
For each one, calculate the exact dollar loss and note whether it looks short-term or long-term based on any purchase date available.
Estimate my marginal federal tax bracket based on my filing status and income.
Tell me honestly whether I have a real tax-loss harvesting opportunity right now, and roughly what it would save me in taxes.
If I said "not sure" about gains, tell me in one or two sentences where I could check (like the Realized Gain/Loss page on my brokerage site or a 1099-B), and give me the estimate assuming no gains as a baseline.
If something is worth harvesting, suggest what to buy instead of the original fund to stay invested without triggering a wash sale (something similar but not "substantially identical" to the fund I'd be selling).
If there is nothing worth harvesting, say so plainly and tell me what would need to change (price movement, account size, realized gains elsewhere) for an opportunity to show up later.
If my filing status is married filing separately, flag whether the lower $1,500 capital loss deduction cap (versus $3,000 for other filing statuses) is relevant to my situation.
Keep the tone direct. I'd rather hear "there's nothing here right now" than have you manufacture urgency where none exists.
Here's what came up when I ran it on mine.
My only taxable holding is actually up 4% right now, so there was nothing to harvest, no loss anywhere in the account. Part of me was hoping to find something to work with, even though a gain is obviously the better outcome.
What the prompt couldn't tell me is the whole picture. It only saw the one account I pasted in, not any other taxable accounts, not my spouse's holdings, not anything sitting in a retirement account where this doesn't even apply. That's true for you too. A clean or a promising result from one account isn't the same as a full check of your household's tax situation.
So treat this as a starting point, not a full plan. It can tell you whether a loss exists on paper and roughly what it might be worth, but wash-sale rules get complicated fast if you hold similar funds across multiple accounts. And if you're sitting on a real opportunity, that's the moment to loop in a CPA or financial advisor before you act, not after.
Think of this as the first pass that tells you whether it's worth digging further, not the final word.
One more thing worth knowing: This is a snapshot, not a one-time check. A loss that doesn't exist today can show up next month if the market dips, and one that exists today can disappear if it recovers.
Two moments are worth rerunning this: after any real market drop, and every December before the tax year closes, since a loss only counts toward this year's taxes if you actually sell before December 31. Same prompt, fresh export, thirty seconds either time.
Keep going? Want to go further? Try these follow-up prompts after you get your initial result:
If you found a loss worth harvesting: "Walk me through exactly how to execute this sale on [brokerage name] without triggering a wash sale, step by step."
If you're not sure what counts as a taxable account: "List every account type I might have and tell me which ones this actually applies to."
If you want the bigger picture: "Now look at my full portfolio across all my accounts, including my spouse's, and tell me if any wash-sale risk exists between them."
If nothing was there this time: "Based on how far my positions would need to drop to create a real opportunity, when in the next 12 months would it actually make sense for me to check this again?"
AI tip worth trying this week: Google AI Mode > AI Overview
I've been using Google's AI Mode a lot lately, and I keep noticing something worth paying attention to: it regularly contradicts the AI Overview that appears at the top of the default search results. Same query, different answers, same platform.
My suggestion: run Google AI Mode by clicking “AI Mode” in the top left rather than just skimming the AI Overview at the top of regular results. You'll often get a more thorough answer, and if something doesn't match what you saw in regular search, that's your signal to dig deeper before acting on it.
Keep up with tech in 5 minutes
TLDR is the free daily email with summaries of the most interesting stories in startups, tech, and programming. The stuff worth knowing, minus the doomscrolling.
Issues are curated by ex-Google and Anthropic engineers and land in your inbox before your morning coffee. A 5-minute read, and you walk into the day already knowing what your team is still catching up on.
Tech is just the start. We also cover AI, marketing, dev, and more. Pick the briefs that match your work.
Free, daily, and read by 7M+ subscribers. Subscribe and let the experts do the digging for the tech news that matters.
One quick note: This newsletter is for educational purposes only and does not constitute financial advice. I'm not a financial advisor — just someone sharing ideas and tools I've found useful. Use what works for you, skip what doesn't, and always do your own research. Some links may be affiliate links or sponsored content for which I may receive compensation.


