Deep67 @billdeep67

$RDDT, highlights from Steve Huffman's Goldman interview: There is a big gap between how well the business is performing and how much better management thinks the product can become. Eight consecutive quarters of 60% revenue growth. Yet the recommendation technology is still 12-18 months behind the state of the art, according to Huffman. THE PRODUCT OPPORTUNITY • Reddit already gets a few hundred thousand app downloads every day that Huffman described as neither Google referred nor paid. • He cited a 50% improvement in new-user retention, helped by better feeds, app performance and moving existing web users into the app. A metric Reddit historically struggled to move is finally showing progress. • His biggest priority is making Reddit better at understanding why you opened it. Looking for advice? A game thread? Something entertaining? Get you there faster. The same improvements can help both newcomers and longtime users. • Core users are engaging and contributing more. Huffman said people generally join more communities and spend more time as they mature on Reddit. New, lighter users can temporarily dilute the averages while expanding the base of potential regulars. WHY THIS COULD BECOME A MUCH BIGGER BUSINESS • Huffman said 40% of Reddit conversations are commercial. Not necessarily people asking “what should I buy?” More often it’s what to wear, watch, play or where to travel. Purchase intent already exists. • That’s the foundation of his advertising argument. He believes Reddit’s interest graph can be more valuable than Meta's social graph. Knowing what someone is researching or considering buying creates opportunities to connect them with relevant advertisers. • He challenged the idea of Reddit hitting a $4-5B revenue ceiling. His reasoning is people don’t stop having interests as they age. The communities they use may change, but Reddit can remain relevant throughout their lives. • The content is also constantly renewing. Users created 1B posts and comments last quarter. Girl Dinner Diaries alone went from zero to 6M readers this year. New reasons to visit keep emerging. HOW TO THINK ABOUT AI AND GOOGLE • An old scrape doesn’t include new discussion. The continuing flow of new experiences and opinions need to be licensed. • Google negotiations are active. • Huffman also separated search traffic from the core revenue engine. The business largely lives in the app and repeat-user base, so SEO swings haven’t produced comparable business volatility. Web visitors still matter as potential future regulars, but not every visit carries the same value. • AI can help Reddit internally while making its human content more distinctive externally. Better recommendations and moderation tools support the product. Original stories, advice and conversations give people a reason to use it. STRONG FINANCIALS • Reddit is approaching a $3B revenue run rate, with 90% gross margins and GAAP profitability. Huffman described the revenue engine as showing “no sign of slowing.” • The balance sheet has $2.8B in cash. Reddit wants to invest in people and compute, make targeted acquisitions, and repurchase stock. • Around $250M has been repurchased this year, with Huffman saying “probably on track for about $500 million.” • Acquisitions are currently about technology and teams that accelerate the roadmap, rather than buying a larger audience. Management’s immediate focus is retaining more of the people already trying Reddit. Overall, strong bullish commentary on the business.

This. If you’re selling $KRKNF / $PNG I promise you will regret it in the next 1-2 yrs. This is the type of trade that will seem so painfully obvious in hindsight. The UUV / XL-UUV market hasn’t even hit early innings.

Kraken Robotics Addict @KrakAdd1ct

5 days ago

$KRKNF $PNG.V 🦑 youtube.com/watch?v=nz4vlS… Anduril aren't quite finished with the production line for the DiveXL, "in the process of completing" at Rhode Island. Currently its only been DiveLD production. building roughly 50 LD vehicles per year, but ramping, they can do 200 a year

3 2 50 11K 11

$RDDT is making the moves I want to see 👌 Underneath the ad Reddit added what they call Redditor Highlights where their AI pulls real conversations about the F 150 and shows them right under the creative. In this one it says Redditors express appreciation for the ride quality power technology and overall driving experience and then it actually links to posts. This is exactly what Reddit needs to be doing and is part of my thesis. For years the bear case on RDDT has been great data but weak monetization and ads that felt random. This format fixes that because no other platform can do it. Google and Meta can target you with AI but only Reddit can show you receipts from a real community of owners who have actually lived with the product for years. It is social proof you cannot fake and it is native to how people already use Reddit which is to check what real people think before buying. It also performs. Reddit said in early tests these conversation summary units lifted click through rate by around nineteen percent compared to standalone ads and their free form AI ads were up twenty eight percent versus other formats. That makes sense because it does not feel like an ad it feels like a summary of what the community already decided. If Reddit wants to close the monetization gap this should be the default not the exception. Let AI automatically match every advertiser to relevant subreddits pull the best community quotes generate the creative and attach the proof. More targeting like this means more relevant ads for users and way higher intent for advertisers. That is how you turn community intelligence into actual revenue and it is why I am still long. (Picture was found on the Reddit stock community)

I think $VST looks compelling here. At $138, Vistra feels like one of the cleaner ways to play the AI/data center power trade without paying a large multiple. The numbers: - Q2 adjusted EBITDA: $1.77B, +31% YoY - 2026 EBITDA guide: $6.8B-$7.6B - 2026 FCF before growth: $3.9B-$4.7B - 2027 core EBITDA opportunity: $7.4B-$7.8B - Shares outstanding down ~30% since 2021 The obvious bottleneck continues to be energy and it seems like $VST is attacking that head on: - AWS signed a 20-year deal for up to 1.2 GW from Comanche Peak. - Meta signed 20-year agreements covering 2.6+ GW across Vistra’s nuclear fleet. Roughly 3.8 GW tied to AWS + Meta alone. Then there is also Helix, the AI infrastructure platform with KKR, NVIDIA and KIA. Vistra is the preferred power provider for this. VST is also acquiring Cogentrix, adding ~5.5 GW of gas generation, with the deal expected to be accretive to FCF/share starting in 2027. The valuation is compelling: At $138: ~13-15x forward EPS ~9-10x 2026 EV/EBITDA ~8x 2027 EV/EBITDA ~10-11x 2026 FCF before growth Meanwhile $CEG trades closer to ~22x forward earnings and ~14-15x EV/EBITDA. I don’t think VST deserves CEG’s full multiple, but the gap looks wide given how much exposure VST has to the same nuclear/power scarcity theme. At the same time, CEG has almost the same amount of debt as VST at $19B. CEG has much more hyperscaler exposure so it deserves a higher premium, but it seems likely for VST to be rerated on solid execution. The risks: - Merchant power prices can weaken - Debt/leverage is meaningful - Nuclear outages/regulatory issues can hurt earnings - AI power demand is already a crowded theme - Heavy hedging limits near-term upside from higher spot power prices - Cogentrix/Helix still need to execute - More generation or slower AI capex could reduce the scarcity premium The risks are important here but they are mainly tied to regulation/execution, the underlying thesis is that energy continues to be a huge bottleneck. What I also like is that you don’t need some crazy AI multiple for this to work and deliver returns from here as the stock is already down 15% YTD: - Bear: $110 7x EBITDA / ~$7B EBITDA - Base: $185 9x EBITDA / ~$8B+ EBITDA - Bull: $240 10-11x EBITDA / ~$9B EBITDA $VST has $1.22B remaining on its buyback authorization and 336M shares outstanding, after repurchasing $778M of stock in 2026 and $6.5B since November 2021, helping reduce its share count by roughly 30% since 2021. That’s actually interesting now because the stock is around $138 so if management keeps allocating FCF to repurchases at these lower prices, every $1B of buybacks would retire roughly 7.2M shares, or a little over 2% of the current share count. Finally, Peter Thiel bought $59M of it in Q2 and David Tepper added to his position by 10% making it now worth 4% of his portfolio with a $350M position size. I took a small starter position today to keep track of it as I'm still doing DD but it does feel more compelling than not.

I have just published a 5,600-word deep dive on $META. Dived into everything from AI capex risks, legal headwinds, opportunities, and valuation. You can read it below; it's free 👇 capitalist-letters.com/p/meta-deep-di…

$RDDT Sentiment at its worst, meanwhile.... -ARPU improving YoY and still at a low base -Midterm ad revenue in H2 -Google licensing deal within next few months -S&P addition -International expansion -No debt, capex -91% GM w/ little incremental capital needed to scale

$RDDT Earlier this year I wrote about precision medicine and Tempus when almost no one wanted to hear it. Basically an echo chamber. Looking back, the pattern is familiar….a structural shift that was obvious if you spent enough time with the business, but largely ignored until the evidence became harder to dismiss. I see something similar setting up with Reddit. I have written about the company in many posts and I want to reinforce the belief because I think the market is starting to set up for exactly this kind of business. This will be a constant theme for me going forward and will probably beat a dead horse here. For the last two years almost all attention and capital has flowed into one trade. The massive buildout around OpenAI, Anthropic and the infrastructure that supports them. It is impressive that it’s gotten this far, but it is built on a model that requires constant spending. Constant compute, constant capex, constant fundraising. That works in a zero interest rate environment where capital is able to flow into. It becomes much more difficult if that capital starts to demand a return. That is why I keep returning to companies that operate on the opposite principle. Companies that do not need to spend to grow to balance out my portfolio. A hedge if you will. Reddit is the clearest example I can find. Its inventory is not made by employees or by studios. It is made by its users. Every post, every comment, every deep niche community in the long tail is created and moderated for free. When the platform adds another ten million users, its content costs do not go up by ten million users worth of content. The incremental cost to serve them is close to zero. It runs with no debt and essentially no capex. No data centers to build, no debt to service. In a potential unwind of the AI infrastructure trade, that leanness is a major advantage. As I have said before, my core thesis remains higher monetization, and I think the path is still very clear. Reddit monetizes far less per hour of attention than its peers. The opportunity is not to add more ads into the feed, but to monetize intent better. People come to subreddits when they are about to make a decision. They are in specific communities asking what to buy, what to use, what to avoid. Better tools for advertisers to target at the subreddit level, better search and discovery inside the platform, and a cleaner and more modern platform appearance for new users would all lift revenue without compromising the core experience. Those are execution problems and not business problems. They are an opportunity. I also want to clarify my view on data licensing. I do not think Reddit needs data licensing to work as an investment. The advertising business alone can justify growth if execution improves. But the data itself remains uniquely valuable and impossible to resource anywhere else. You cannot recreate twenty years of authentic human debate, troubleshooting, product comparisons and lived experience by scraping the web or generating it synthetically. Human judgment is one of a kind. Whether the company licenses it heavily or not, the scarcity of it only increases as synthetic content floods everything else. Finally, I think the continued erosion of trust in platforms like Meta will help Reddit more than people expect. People are tired of being optimized for time spent. When the feed is purely algorithmic and the content feels increasingly synthetic, people start to look for a place where discussion is intentional and driven by interest rather than identity. That is exactly what Reddit offers. It does not need to replace Meta. It just needs to absorb the moments when people want a real answer from real people.

@TheWiseIC Thanks for the recorded video! 👍🏻! Will you be selling Covered calls on youre position ?

@TheWiseIC Definitely wanna get this recorded version please post when done gotta fly!! 👍🏻

1/ muse spark 1.2 is a very strong multimodal model—it can do visual coding, robotics planning, and audio-visual understanding that all come together through agentic tools.

Reddit isn't cited at all anymore. Nope, not at all. LOL

This is genuinely the best post I’ve ever seen on an individual stock. $CIFR is next.

Coffee House Stocks @CoffeeStocksGuy

3 weeks ago

x.com/i/article/2089…

25 40 325 604K 402

Now that $RDDT is going into the S&P 500 next week, the index funds only have 2 days to buy ~30.4M shares which is ~21.5% of the float… curious to see where the price goes over the next couple days but I suspect it goes higher on forced buying. Feel free to check my numbers but here’s how I came up with it… $RDDT’s float is approx 141.6M shares, multiply that by $158 (last nights close) and you get a float adjusted market cap of $22.4B $22.4B market cap would be ~0.032% position size in the S&P 500 There’s approx $15 trillion indexed to the S&P 500 and those funds will need to buy $RDDT. $15 trillion x 0.00032 = $4.8 billion $4.8 billion / $158 = 30.4M shares 30.4M shares / 141.6M float = 21.5% NFA. DYOR. *We are long $RDDT at @FirstWaveFund

Jonah Lupton @JonahLupton

4 weeks ago

Price action on $RDDT could get wild the next 2 days… 13.6% short interest and according to a friend the $SPY index funds only have 2 days to do their buying and $RDDT doesn’t feel like the kind of stock where shareholders will be selling into this pop because we know there’s a

keyboard_arrow_left Previous keyboard_arrow_right Next

27 11 217 90K 32

$META CEO Mark wrote a piece laying out his vision for “personal superintelligence,” with some implications for AI products, compute spending, open source strategy & the future of work. Here are some of the most important parts if you don’t have time to read the full piece: “Everyone will have an exceptionally capable personal agent that understands you, your goals, and everything you care about. Your agent will work 24/7 on your behalf to improve your relationships, health, career, finances, home management, hobbies, and more.” “You'll be able to interact with your agent through any device, including your glasses to keep you present in the moment with the people you care about.” “We will offer free versions that will be accessible to billions of people. For those who want to pay to use more compute, there will be a dynamic auction mechanism that will guarantee that everyone gets the lowest price possible for the intelligence and compute they're using while also ensuring the capacity is used for whatever people collectively find most valuable.” “Company sizes may shrink -- just as they did in the transition from industrial giants to tech companies. But this doesn't mean fewer jobs overall. It implies a larger number of companies with fewer people each.” “I expect we will start seeing small numbers of people with personal superintelligence agents able to run companies at significant scale.” “We help keep electricity prices low by building our own energy-generating infrastructure wherever we invest.” “No matter how intelligent AI becomes, there will always be a finite amount of compute and therefore an opportunity cost for how we use it.” “Now that Meta Superintelligence Labs are up and running, we will resume releasing some open source models soon.” “For the US to lead in open source, we will need to rethink our policies in several areas, including distillation and data use in training.” “Countries like China are bringing online 1GW+ of nuclear capacity every other week, so we will need to accelerate building both energy and data centers to remain competitive.” “Any policy that slows American model releases -- even by a month -- could add significant risk to American leadership while letting foreign models race ahead.” “I propose that companies developing frontier AI should commit significant technical resources towards helping the government harden critical infrastructure. I also propose that frontier AI labs should share intermediate training checkpoints of new models for government use and review rather than waiting until training has completed.” “Meta, other frontier labs, and clouds committed to giving people a greater balance of power must collectively build out a sufficiently large amount of compute such that we can allocate enough to recursive self-improvement to remain competitive while still committing the significant majority towards people's individual goals.” “For example, if a self-improving AI system focused on optimizing its compute efficiency, it could theoretically invent ways to squeeze 100x or more intelligence out of each gigawatt.” “At Meta, we will focus our efforts on delivering personal superintelligence to billions of people and small businesses to help everyone achieve their goals and invent the things they want to see in the world.” “The future will be created by all of us and a reflection of all of our values and perspectives.”

Mark Zuckerberg @finkd

a month ago

I believe everyone should have access to superintelligence, and I wrote a long piece about Meta's philosophy and values for building a positive future for everyone. meta.com/thefutureisfor…

3K 2K 19K 4.3M 6K

@bjmtweets @steady_profits Any amount of shares to sell CCs juicy prem in reddit ? @bjmtweets

Yeah, I know, another long ass article about $RDDT . Here’s the short version: I’m buying the dip, and Ken Griffin will buy my book once I’ve short squeezed him.

Kenshii @KenshiiWCWC

a month ago

Market freaked out over 300k sequential U.S. users. Meanwhile ads +64%, U.S. ARPU +51%, FCF more than doubled, and the data business is still barely priced. $140 feels wrong for $RDDT. As promised, I'll give you my fair and honest opinion. x.com/i/article/2084…

0 0 8 3K 1

$RDDT is close to $META P/E on an annualized basis now. And Reddit is growing DAU and WAU much faster. Meta is only growing DAU 3% YoY. Oh, and Reddit also has much more room for ARPU expansion. My best guess is the DAU story is to shake out retail before more institutional accumulation. This business is too high margin and early in monetization stage. Reddit is the highest quality business I’ve found with a NTM P/E under 20.

@Mindset4Money_X Do you have a position?!!

@GauchoRico No addition to RDDT???

@TemptInvest Well said!