Hey all, it was an up and down month for my portfolio. Fortunately a few companies held up well like IREN, ELVA, and SKYT.
I had a breakthrough with setting up a new screener this month that uses analyst estimates as a way to estimate the company’s guidance. There is a video on my Youtube channel describing how this screener works if that sounds interesting.
Returns are,
2024: +146%
2025: +112%
2026: +4% YTD
Cumulative: +442%
Allocations are,
Iren IREN - 18.8%
Astera Labs ALAB - 16.1%
AppLovin APP - 14%
Electrovaya ELVA - 12.5%
Figure Technology FIGR - 9.2%
Paymentus PAY - 3.9%
Credo CRDO - 3.7%
Pattern Group PTRN - 3.5%
Reddit RDDT - 3.3%
Dave Inc DAVE - 3.3%
Shoals Technologies SHLS - 3.1%
Micron MU - 2.6%
T1 Energy TE - 1.9%
BioHarvest Sciences BHST - 1.4%
SuperMicro SMCI - 1.2%
Duos Technologies DUOT - 1%
Motorsport Games MSGM - 0.5%
Promising new ideas on the month,
Accelerant Holdings ARX - insurance and risk marketplace
Energy Vault Holdings NRGV - utility scale energy storage
Zedcor ZDC.V - AI enabled surveillance and 24/7 monitoring
Anaergia ANRG.TO - waste to energy platform
Blaize Holdings BZAI - AI chip maker for edge data centers
Figma FIG - SaaS for product management and design
Cloudflare NET - Security platform for protecting apps
Sandisk SNDK - Old school memory maker getting a boost from AI
SiTime SITM - timing devices for semiconductors
Nanya Technologies 2408.TW - DRAM maker gaining share in Taiwan
I am starting to get ready for the earnings season now with SuperMicro and Reddit reporting earning this week.
Best of luck to everyone with their portfolio this year!
Figma is mostly a UI design, app prototyping and deployment set of tools. I don’t know what actual product management tools they have. They agreed to be bought by Adobe a couple years ago, but the government nixed the deal since they felt Adobe would be too monopolizing in the design space.
I was looking at their numbers this morning. It isn’t growing at a super growth rate, but might be worth looking at as Adobe is also at a 4 year low today. If they are taking share, it could be a good investment. The issue is that I think everyone is expecting ai (like $GOOG’s Nano Banana) taking this market over. Numbers look decent though.
The last couple engineering jobs I had were both using Figma to as the main product development tool. At least what I saw is that it’s become the de-facto standard for building mobile apps from the product side. Typically if there was a new feature the designers or product will produce the screens and transitions in Figma and it stores all the underlying primitive objects that are needed for iOS or Android development. Once the designers have finished up, they will often look for feedback from engineers, demoing within Figma. There are all sorts of integrations as well such as attaching the design to engineering tickets in Atlassian’s JIRA.
I also saw how when there was a presentation to executives, what used to be done in a powerpoint is now done through screen-share and Figma. Since the whole app is designed through Figma is just makes sense to use the app natively like that rather than build a presentation. Overall, I see this as a product that is used across a business by a lot of different roles, whether it’s engineering, design, product, or management.
It seems like Figma is getting scooped up in the narrative that AI will completely disrupt their business. My take at least on Figma is that it will be fairly challenging to disrupt them now. Figma helps lock down a precise definition of how the app will work working with the tooling that iOS and Android have. Those tools like Nano Banana or ChatGPT can produce one off designs that are not really useful in the engineering sense yet. Of course this space is evolving fast though but I think it’s going to be hard for pure AI to replace Figma.
From the investing side, what I’m really interested to see with Figma is if they are reaching GAAP profitability on the next quarter. Their first posted quarter was 52M of GAAP net income, but the next quarter had all the IPO expenses jammed in there. Analysts are projecting a quarter that is heavily GAAP negative, but I’ll be kind of surprised by that outcome. This seems like a fairly light business model in terms of SaaS. I do not think this company needs a huge salesforce as their product is well known from the ground up.
I was referencing the Net Retention for customers over $10k being 131% and the fact that they guided up and have beat by 2-5% points typically. Thanks for your thoughts as well.
As a side point, are you looking at GAAP net income numbers specifically because of the market temperature? I think that was definitely one of the lessons from 2022 that some of us pulled forward. The companies that tended to be growing AND had a P/E obviously did better back then (as I think you know and might be applying).
Yes, my teams also used Figma at the previous 2 companies I was at. It was really great (probably still is), because you can use it not only for static “here’s what the app will look like” stills, but you can also make it live in demo/prototype form in minutes, and then when you’re ready for the real deal, what you did in Figma carries over as well to the final product.
I’m not sure what I see there makes a big difference, but I’m not that close to it anymore. I suspect it’s somewhat like graphic design - today if you’re just doing something internal, like for a presentation, AI can do the graphics for you probably better than you ever had. But, it’s still not really ready for an ad you’re going to place on bus canopy walls. I would think having AI produce Figma files might be the best of both - get started quickly, but then have human-tweakable and then production-ready assets as output.
Figma’s products have been great - if they’ve got a good handle on monetizing the business, it’s definitely worth a look.
I have been reading and watching plenty about AI for Devs and it looks like most AI models are heavily trained to use Figma. Looking at OpenAI most recent release Codex App has 2 different skills to use Figma (MCP and Design).
The release video from OpenAI has Figma as one of the three installed skills. 4 minute long video but you can verify at the 2:41 time mark.
My experience using AI has made me very skeptical of SaaS companies but I think several SaaS companies will become larger winners in the future as AI funnels people into some SaaS products. Right now it looks like they are training the AI to use Figma. But if AI capabilities continue to grow at this pace it might move Figma into the camp of SaaS that struggle to compete against AI in the future.
They just announced good earnings, for those that believe them:
Adjusted EPS: $0.69 vs analyst estimates of $0.49 (41.4% beat)
Adjusted EBITDA: $592 million vs analyst estimates of $440.9 million (4.7% margin, 34.3% beat)
Revenue guidance for next year is $40 billion (midpoint) from $36 billion, a 11.1% increase
Adjusted EPS guidance for Q1 CY2026 is $0.60 at the midpoint, above analyst estimates of $0.52
Market cap was almost $18B before the announcement.
However:
• Operating margin down to 3.7% from 6.5% from same quarter last year.
• Gross margin also declined.
• Customer concentration: One customer accounted for 63% of revenue for the quarter! Clearly there’s a hyperscaler using SMCI heavily.
I admit the story sounds OK, but adding the margin pressure (traditionally an issue for this kind of business) and the single customer concentration on top of what has historically been fraudulent or at least iffy accounting practices, I’m not particulary interested in getting back in.
I looked at SMCI again as well. The revenue gorwth story is great, but margins are terrible. I decided against even a small position, and I wasn’t even aware of the customer conentration. Given all that, the stock popped today - we’ll see if it’s sticky are quickly sold off.
I view GAAP profitability as a proxy for the business becoming self sustaining. This allows the company to accumulate cash rather than burn through cash. In turn this cash generation helps improve the value of the stock. I’ll compare and contrast two companies in SaaS like Cloudflare and Paymentus. Both companies are FCF positive with NET having 83M of FCF quarter and Paymentus have 35M of FCF. The cash flow indicates both businesses have no liquidity issues, yet the difference is in the bottom line.
Paymentus’ cash balance over the past over the past year has gone like this, 188 → 206 → 246 → 266 → 288. This gives the company flexibility to invest further while improving their valuation. Cloudflare on the other hand had to raise 2B in convertible notes because their revenue does not translate to the bottom line, and thus they are raising capital from outside sources for expansion of their business.
Good points here, I’m also getting further away from having hands on experience with the product as it has been over year since I’ve used it. The AI space is moving fast as well and it’s hard to know when there will be some sort of tipping point on AI tools. I have seen from a number of different places in the past week talk about scientist’s viewpoint on AI. The AI has solved multiple math and physics problems that were unsolved by humans, and its accuracy being set on these types of problems is improving.
Probably part of the reason for the selloff is that the market is thinking years down the road for how these businesses may get disrupted. I don’t see this pure disruption scenario happening in the next couple of quarters, but who knows how the landscape will look in a couple of years.
This investment in Supermicro was about as close as I get to a trading position with my strategy. I’m in agreement there is way elevated accounting risk still, along with poor margins, and throwing on top of that a customer concentration issue.
Where I saw an opportunity this quarter is that the stock price had gotten too beat up even factoring in high probabilities of recurring accounting issues. Heading into the earnings the market cap was 18B with a projected revenue of 10 - 11B by management. That includes them saying they had a 13B order recently, management saying they gave a very conservative guide, and are getting up to a 100B annual run rate soon.
Basically the market is pricing in an incredible amount of risk with the current valuation. A lot of times when this scenario comes up a company just meeting expectations can drive the stock price because the market views the stock so pessimistically.
I am not too interested to make this a long term holding right now because I see companies like ALAB, CRDO, MU, IREN, or SNDK having better propositions. Especially MU has a big contrast in pricing power on their side, while pricing power is going against Supermicro.
Supermico does have a new buildings block product though which is over 20% margin and making up 4% of revenue. They say this piece is expanding, but on gross margins the management said they expect it to only go up 30 basis points next quarter. One last thought is that even though the margins are tiny, we are talking about a massive top line number that still translated to 401M of net income. I will likely take another look at Supermicro again before the next earnings to see if the risk/reward again looks attractive heading into that report.
