The Sleep Well Portfolio consists of time-tested, market-leading companies. We screen them with a rigorous checklist and regularly track their theses to determine when to buy. So far, we have made one mistake, but are happy with the current 13 holdings. All about us here.
Hi all,
A few days ago, I published the 25th and 26th Sleep Well picks that I believe will benefit tremendously from the AI 5-10 year buildout. The current portfolio is overexposed to industries that aren’t likely to change much in the next 5-10 years, so these two add growth while balancing that with a resilient business run by veteran management, and they're available at a reasonable valuation, in my opinion. What I like most about these two picks is that, amid AI excitement, I won’t have to guess which leading-edge chips, memory, or AI models will win for them to win. Why?
These businesses are simply the AI picks and shovels; they provide irreplaceable machines and components to build the digital and physical AI infrastructure. As the installed base of these machines grows, they also generate more predictable earnings that help dampen the cyclicality of the industry. I have a feeling I will like them as much as my first AI/semiconductor picks and shovels - The VAT Group.
As a result, I placed an order for both, and prices are now very close to being filled for the Sleep Well Portfolio. Check them out.
Please note these are newer picks and potential positions; I’ll buy slowly and add more details as we get to know them better. The deep dives are comprehensive, but we often only truly understand the stock/business once we've owned it for a while.
Trimming CrowdStrike - Security layer of AI
At the August and September review, I put CrowdStrike as a candidate for a trim, given the high expectations baked into its 140x free cash flow valuation and $270B market cap. Prices have been very volatile since, and the current range is comparable to the peak COVID period. The difference is that ARR was growing at >100% vs. 25-30% today, and the dollar-based net retention rate was 125% vs. 115% today.
AI has fundamentally changed the threat surface, and CrowdStrike has upgraded its long-term guidance to $20B ARR by 2035, a near 4x in 9 years or a 15% CAGR from $6B ARR today. However, even with a very optimistic exit FCF multiple of 40x (yr 2035), it leaves very little money to be made at $270B today.
Quick math: Free cash flow by 2035: $7B (35% margin of $20B) x 40x = $280B market cap. (vs. $270B)
I have done a longer commentary on the state of cybersecurity and CrowdStrike in the past few months, and now I feel ready to trim more shares of CrowdStrike. I’ll send you a note once the order gets filled.
Portfolio summary of Oct 2026
Here is how the portfolio looks today:
Since making the grave error 12 months ago, I have been more selective with my picks and my buys. This period also trained me to be a little more patient, cause the the hardest thing in investing is waiting for the ‘reasonable’ moment to buy/sell, and even harder, buying slowly when everyone is selling.
I am referring to Sea Limited right now. The share price is hovering at levels where I bought significantly in 2023. A valuation of ~11x EBITDA is at a record low. Meanwhile, the three segments are all doing very well (growing profitably). Yes, Sea Cloud is giving investors a lot to imagine, since the announcement left very few details, but knowing Forrest Li, I don’t think he'll make the same mistake twice (burning cash on lines of business where competition is already strong). I am excited for the development of the 4th revenue pillar and will be all ears for Q3’26 earnings in two weeks. For now, you can read my thoughts on the big announcement here.
Similar to Sea, Meli and Wisetech are also having a tough time (stock-wise), despite the business doing well and should benefit from more AI deployments.
So going back to ‘buying when everyone else is selling,’ as the hardest thing in investing, the caveat is you have to be comfortable with the business that you own, know your financial goal, know what is enough—all the things that help you distinguish between the trajectory of a stock and a business, because, in the short run, they can be completely uncorrelated. But over the long run, I believe the business performance will always be the north star.
As always, time will tell. I have survived the 2008, 2015, 2020, 2022, and 2025 down years. Market cycles are a feature of long-term investing, not a bug.
By 2037 (when my daughters will have full control of the Sleep Well Portfolio), none of this would matter. I continue to put aside some savings for them and own more winners - ‘winner keeps winning ’- in both bear and bull markets.
If you are new to SWI, check out our FAQ and Owner Manual.
If you are already an annual member, you can access the Sleep Well Portfolio and Thesis Tracker via the link below.
