Wpr101's November 2025 portfolio review

Investment Analysis Clubs Saul’s Investing Discussions
1

Hey all, like many others it was a tough month for results. I cut a few smaller companies from the portfolio and added three new ones this month with SkyWater Technologies SKYT, Figure Technologies FIGR, and Evaxion EVAX.

This month I decided not to label the smaller companies “Microcap Ventures” because a few companies moved up in allocation to be normal positions, and a number of others disappointed which were sold. However, the basic idea stands that newer lower allocation positions are likely to get less attention. I also added a section on market commentary to the video this month detailing some thoughts on dealing with downturns in growth stocks.

Posted results at the end of November,

  • 2024: +146%
  • 2025: +104% YTD
  • Cumulative return: +402%

Allocations at the end of November are,

  • AppLovin APP - 20.5%
  • Astera Labs ALAB - 19.8%
  • Iren IREN - 17.8%
  • Electrovaya ELVA - 11.2%
  • Hive Digital Technologies HIVE - 9.6%
  • Reddit RDDT - 4.6%
  • Duos Technologies DUOT - 4.0%
  • Dave Inc DAVE - 3.9%
  • BioHarvest Sciences BHST - 3.5%
  • Paymentus PAY - 1.4%
  • SkyWater Technologies SKYT - 1.4%
  • Figure Technologies FIGR - 1.3%
  • Evaxion EVAX - 0.6%
  • Organigram OGI - 0.6%

Promising new companies this month included six new companies and three companies we used to own that have piqued our interest again,

  • Exzeo Group XZO - IPO spinoff for insurance SaaS platform
  • Organogenesis ORGO - Biotech for skin substitutes, competitor to BioStem
  • Telos Corporation TLS - SaaS security solutions primarily sold to the government
  • Daily Journal Corporation DJCO - newspaper business with SaaS offering
  • VusionGroup VU.PA - French IOT, cloud and data analytics business
  • Adaptive Biotechnologies ADPT - genetic and clinical diagnostic solutions
  • GeneDx WGS - genetic testing expanding to NICU + adult
  • Amprius Technologies APMX - solid state battery company with business momentum
  • Celestica CLS - hardware provider selling to hyperscalers and expanding to ASIC solutions

Any feedback on the holdings, new ideas or strategy is always welcome.

59 Likes
2

Daily Journal Corporation DJCO - newspaper business with SaaS offering

Well bless my heart, a Charlie Munger company that fits on Saul’s board. This is not an every day occurance by any means. I haven’t looked at DJCO in a few years but I’m certainly going to dive in now.

12 Likes
3

Regarding Skywater, I am skeptical. Just some quick background: I am very familiar with the semiconductor industry, having worked in it for 25 years at various roles (engineering, marketing, sales, etc.) and companies. And I am quite familiar with Skywater as several of my former colleagues now work there; furthermore, I have had several discussion with Skywater regarding available semiconductor technologies, foundational IP (e.g., memories, I/O pads, standard digital cell libraries, etc.) and even procured some quotes from them for their foundry offerings.

First off, understand that Skywater is only a semiconductor foundry, a “fab”. They manufacture chips, but do not have the resources to define and design new chips for the market. They are “just” the manufacturing piece in the industry–similar to TSMC. Skywater is the complete opposite business model of “fabless” companies like ALAB or NVDA who design and sell microchips but rely on a third-party fabs. Both are viable business models, but the later tend to get (much) higher valuation multiples simply because they do not carry the manufacturing overhead.

It does seem that Skywater’s business is picking up, but they have historically struggled to fill their fabs since they were spun out of Cypress a few years ago. They had contracts to keep selling some Cypress products but ultimately have to find their own customers and new business–and it typically takes 1-3 years to design, prototype, and ramp a new microchip. Their initial strategy was to go hard after the Aerospace and Defense market because they are a US-based foundry. I don’t believe this has panned out too well as one of my former colleagues (a director of a related business unit) was just let go.

Please note that Skywater did not acquire Infineon (a $46B market cap company), nor was the price $350M for the piece they did acquire. The $350M figure is a new loan Skywater procured from several lenders, the majority of which is still untouched. The piece Skywater acquired is one “older” 200mm wafer fab for ~$93M ($73M at close, ~$20M due later). I say “older” because the Infinion fab still has new technology (65nm) than any legacy Skywater fab (130nm & 90nm), but still much older than the cutting edge technologies used for NVDA/HPC chips (e.g., 3nm). But not all chips need the processing horsepower that HPC needs so there should be plenty of opportunity to use the newly acquired fab (as well as the older ones).

Finally, I am not convinced that quantum computing is going to move the dial anytime soon for Skywater. Quantum computing is still just early and we are barely at the “proof of concept” phase and (as Jensen Huang pointed out almost one year ago) we are 15-30 years from useful quantum computing. Also, there are 3-4 very different technological approaches to realizing useful quantum computing (e.g., room-temp photonic, some cryogenic supercomputing methods, etc.). At my current company, we did a short feasibility project for a well-funded quantum computing company and they told us they can currently produce 100s qubits but they need to be at 10,000s of qubits to be viable. That’s two orders of magnitude of improvement needed! Even if the progress was following Moore’s law (which it isn’t), they would need 10 years to get there. So yes, 14-29 years seems about right to me. It would be very instructive if Skywater reported the amount of revenue associated with quantum computing as that would tell the true story. Until then, it is just speculation.

For many reasons, I really hope Skywater does well. But I do not see the drivers needed to be a true Saul stock for a sustained period. For anyone invested, I hope I’m wrong!

45 Likes
4

@7and1 That behind the scenes details on SkyWater and the semiconductor industry is super helpful!

I am definitely in agreement here that SkyWater does not present the same level of innovation that an Astera Labs or Nvidia has. It seems more like a manufacturer of other’s designs through their fab, although it did sound promising that they can do “technology as a service”, to implement a whole variety of quantum designs along with more regular semiconductor designs. I see the valuation as reflecting this, where the run-rate P/S is around 1 for SkyWater and around 25 for ALAB or CRDO.

I thought this aspect of catering to defense and customers who want US based only for various reasons may benefit SkyWater. With the current environment of tariffs, this seems like a company which could actually benefit from the situation. I was wondering about their claim that they are the only “pure play” fab in the US like this, or if that is marketing spin.

Wow, looks like I got those details completely wrong! It probably stems from my initial research using AI on the acquisition that I misunderstood how the acquisition worked there. Appreciate the detailed breakdown there.

This aspect of whether the quantum business will be enough to drive the needle seems important for the investing thesis. I noticed they never gave any numbers on what portion quantum makes up for them, leading me to believe it is small. I have seen companies do this before where they emphasize the growth of a new product with stunning qoq numbers but it turns out the base is low.

I’m looking at these companies which are quantum stocks and the valuations seem nuts to me. Here’s a quick glance,

  • IONQ, 40M revenue, -1.06B net income, market cap 17B
  • QBTS, 4M revenue, -140M net income, market cap 7.9B
  • QUBT, 0.4M revenue, 2.4M net income, market cap 2.5B
  • RGTI, 1.9M revenue, -201M net income, market cap 8.1B

It is interesting why the market is paying up for all these companies which are practically pre-revenue, especially if the payoff is 10+ years down the road. However, I can see SkyWater benefitting from creating designs for quantum companies like these and other startups, even if the final product is not selling or doing well yet.

I mentioned in the video SkyWater might have the potential to get considered as a quantum stock, but I don’t want to factor that in too strongly.

Overall, I agree with your sentiment this company is not fitting the exact mold of what would make a great Saul stock. Mostly this is because of the acquisition and we do not have accurate comps yet for yoy or organic growth. Additionally, we don’t have details on the amount of quantum revenue.

On the other hand the stock is priced quite low especially if they are able to grow at all. Even a small amount of quantum revenue may be enough to start inching them up to beat expectations. Their last quarter came in well ahead of guidance and analyst expectations. Also my subjective take is the management was a bit surprised by their outsized success in the quarter.

16 Likes
5

Nice, yea this really a fascinating and unique opportunity in the market right now. The 443M of marketable securities or basically a stock portfolio makes this company quite a bit different than what we would usually analyze. That likely provides a floor to the stock price, and there could be some interesting arbitrage opportunities, although that investing style is a bit different.

I also noticed that DJCO industry is listed with the sector as Information Technology, and the industry as software. It’s possible the software aspect is not as hidden as I mentioned in the video. I’m kind of surprised the company has not rebranded on the title because it sounds like a newspaper on first glance.

It would be nice if the company did earnings calls though. I do find it bizarre that a company which obviously likes stock investing would go out of their way to do only the bare minimum of filings. The filings provide a lot of information but it is a pain to parse out all the boilerplate language to find the main details regarding the company.

8 Likes
6

Some news on IREN over the last two days. One is that it was included in Dan Ive’s AI 30 Winner’s List (as well as Coreweave). And this…

IREN Limited recently announced a $2 billion offering of convertible senior notes, split into two $1 billion tranches due in 2032 and 2033. The company is also conducting a concurrent equity raise to repurchase existing debt, as part of a strategy to optimize its capital structure to support its AI infrastructure and data center growth.

Details of the New Debt Program

  • Total Offering Amount: Up to $2.3 billion in new convertible notes (including initial purchasers’ options).

  • Structure: The offering consists of two separate private placements to qualified institutional buyers:

    • $1 billion in convertible senior notes due June 1, 2032.

    • $1 billion in convertible senior notes due June 1, 2033.

    • Initial purchasers have an option to buy an additional $150 million of each series within 13 days of the issuance.

  • Key Terms: The notes will be senior and unsecured, pay interest semi-annually, and include conversion, redemption, and repurchase rights.

  • Purpose: The primary goal is to manage IREN’s existing debt profile by using the proceeds to repurchase portions of outstanding 2029 and 2030 convertible notes. Proceeds will also be used for capped call transactions (to reduce potential share dilution) and general corporate purposes.

  • Strategic Context: This move is part of IREN’s strategic pivot from Bitcoin mining to high-performance computing (HPC) and AI cloud services, following a significant deal with Microsoft.

18 Likes
7

I’d place that claim more in the marketing bucket. There are several direct competitors to Skywater that offer onshore fabs in the US: onsemi, XFAB, Global Foundries, Jazz Semi Trusted Foundry, LA Semi (but might fold soon), Microchip, and even TI and Intel are offering foundry services. Also, TSMC now has a US-based Fab in Arizona (and building two more).

Granted, these competitors offer a wide array of semiconductor technologies that have varying degrees of overlap with Skywater’s offerings. But “pure-play” merely means they only offer fab services, unlike a company such as onsemi which offers fab services in addition to creating and selling their own products (i.e., an integrated device manufacturer, or IDM).

From the perspective of a fabless company that is seeking foundry services, there is no real value add to use a pure-play fab. The deciding factor is what technology offerings (and cost) meet the chip (and market) demands.

I will point out that most fabs are not willing to custom tailor a semiconductor process to meet the needs of a specific customer. Clearly, Skywater is doing this for at least one quantum computing company. If that business takes off, it could be an advantage.

20 Likes
8

I would imagine the concentrated equity portfolio protects against downturns but also would be a drag on any gains due to the software business. So probably not really a ‘Saul stock’. Probably of more interest to value investors, many of whom fled MF after the platform change; they’re all hiding out at a site styled after the old MF site. If you’re still looking into the company, here’s a thread with some background information that might be useful: Post #11684 by EVBigMacMeal on the Berkshire Hathaway board

4 Likes