Consensus Long @ConsensusLong

@aleabitoreddit Don’t forget about $CLS highest exposure to $GOOGL out there!

Asking all the smart people on Twitter $MU $SNDK $DRAM Came across this chart from UBS that shows memory is going from 13% of hyperscale capex in 2025 -> 33% in 2026 -> 57% in 2027. #1 - Is this accurate / true? #2 - how is this sustainable? $GOOGL $AMZN $MSFT have some of the smartest engineers in the world, surely their #1 priority will quickly become figuring out ways to reduce memory costs!

$HUT adding another $10B deal with a second IG hyperscale tenant at a second location! Despite the +165% move in the stock 🥳over the last 4x months there is still lots of room to go. The second two phases of $HUT River Bend site is worth another $124/share and the second phase of TX site (today's announcement) worth another $45/share!

Consensus Long @ConsensusLong

6 months ago

$HUT signed a $7B revenue, 15 year agreement (with additional 15 year extensions) with Fluidstack (fully backed by $GOOGL AA+ credit rating) who will be renting capacity to @AnthropicAI This deal is one of the best infrastructure deals I have ever seen: 1/ $HUT set a new high

1 0 0 493 1

Whether it’s existing consulting firms, new ones that emerge, FDEs from agent vendors, or new internal agent engineering roles, the amount of work that is going to be created to implement agents in enterprises will exceed anything we imagine today. The complexity of implementing agents in any existing organizations is very real. When I talk to large enterprises, as you move from a chat paradigm to agents that participate in meaningful workflows, there are a number of things they need to do. First, you have to get agents to be able to talk to your data securely across your systems. In many cases, enterprises have decades of legacy infrastructure that contain the valuable context for AI agents. That’s going to take a ton of work to go modernize and move to systems that work well with agents. Then, you need to ensure that you’ve implemented agents with the right access controls and entitlements, the right scopes to be safely used, and have ways of monitoring, logging, and securing the work that they do. Next, you need to actually document the processes in the organization in a way that agents can utilize for doing the work. You also need to figure out what the new workflow looks like when agents and people are working together on a process, and who steps in where. Just replicating the old workflow will mute the gains. Oh and you likely need to create evals for your top new end-state processes. Finally, you have to keep up with a rapidly changing set of best practices and architectural shifts happening in the agent space. While it’s fun for people to change their personal productivity tools on a dime, it’s 100X harder to do this in a business process. The speed of change is a blessing and a curse right now for anyone trying to keep a stable system design. All of this means that individuals and companies that develop expertise on the above set of components (and more) are going to be needed to help organizations actually implement agents at scale. This is also the rationale for vertical AI agents right now that can go in deep on a business domain and help bring automation to it. This is a huge opportunity right now whether you’re doing this internally or as an external business provider.

@chamath $HUT $APLD $WULF $CIFR

@paulswaney3 Application

@RHouseResearch $HUT has a similar deal with them as well!

@JonahLupton But bro didn’t you know that North American logged in DAU growth is going to be 200k this quarter and the street has 300k + these deals won’t be announced for another 4-5 months… clearly a funding short!

@buccocapital Customer bases got hollowed out. Big chunk of business from start-ups doing drug discovery and running early trials - VC funding dried up for the cohort in 2022

@HedgeyeTech Bro the BTC miner thesis is literally BTC mining is a $0 and they are converting to DC developers. Most of these guys have successfully made the transition

I had the opportunity to meet with $RDDT in a large group meeting for institutional investors this week. RDDT is a stock I have owned since the IPO and I have been consistently pleased with their execution. RDDT’s business is firing on all cylinders. Advertising revenue grew +75% last quarter – growth was broad based across all verticals, geographies, and advertiser objectives. Growth is being driven by a variety of levers – primary user and engagement growth (high quality growth) and to a lesser extent increasing ad load and higher ad pricing. The company positively highlighted the traction they see in performance advertising, which is over 50% of total advertising revenues and growing >100% YoY. To be fair, RDDT includes both mid-funnel (lead-gen type advertising) and true bottom the funnel (immediate conversion type advertising) as performance advertising. My channel checks point to significant strength in mid-funnel with true bottom-funnel direct response still being in the early days. RDDT management spoke optimistically about the traction they are seeing with true direct response ad units, like dynamic product ads and shopping ads, and were excited about the rollout of app download ads in 2026. The bottom line here is that growth and momentum appear to be highly durable. The company runs an extremely tight ship – one of the best teams in the sector (I know it’s a low bar!) when it comes to expense discipline and capital allocation. EBITDA margins expanded to 38% in 2025 from 23% in 2024 and the company has been profitable on a GAAP basis for six quarters in a row. During the meeting management pointed to more of the same – we should expect 50%-ish incremental margins going forward and RDDT is heading towards being a 50% EBITDA margin company over the medium term. For 2025 RDDT scored a 108% on the rule-of-40% score, one of the best in the market. The one rub on the story is around user growth. After some turbulence in 1H25 RDDT is on more solid footing but trends are still just okay; total daily users grew +19% in 4Q25. Management’s continues to highlight the significant opportunity they have to convert their 470M+ weekly users to daily users. They also made it clear that they are prioritizing growing user engagement and time spent on the platform by allowing people to interact with communities and engage with other users without forcing them to log in. I think that this is good for the long-term health of the business and good for current advertising trends, albeit at the expense of the KPI that short-term investors are most focused on. Unfortunately, RDDT’s stock has not performed anywhere close to its business performance. The company’s stock is down -44% YTD and the stock is down by more than 50% relative to the all-time-high it made in September 2025. I struggle to find a reason for its terrible performance, especially since it never traded at an egregious multiple to begin with. I guess one can point to US “logged-in” DAUs falling -100K QoQ (less than 1%), but its hard to see how that can take away from a company that saw forward EBITDA revised up by 60%+ over the past year. Currently RDDT trades at 25x 2026 free cash flow and 16x 2027 free cash flow, an absolute bargain compared to the broader market before you even factor in RDDT’s best-in-market growth rate. Its very rate that the market gives you the opportunity to own a rapidly growing, highly profitable, network effect business model at such an attractive price. Below is a picture of RDDT’s stock price (white line) versus RDDT’s 2027 EBITDA estimate (blue line). In my experience jaws like this always close – given my expectation for 2027 EBITDA to continue to be revised up, it is highly likely this resolves itself with RDDT moving meaningfully (100%+) higher.

x.com/i/article/2017…

Great analysis! Any idea how the compute usage (MWs) splits out between training and revenue generating activities? The hyperscalers/neoclouds make $10B/GW renting out GPUs. This is the COGS for anyone sitting on top of their infrastructure (everyone) so revenue/GW needs to grow and be a lot higher than $10B/GW for the model layer and application layer to work

@JaredSleeper Great survey work! Seems like cutting SBC is an easy win, yet none of the companies pulling the lever. Market is doing it for them… making the value of that SBC worth less and less everyday 😂

Great advice from someone who has been there and done that in a (very) big way!

Henry Schuck @HenryLSchuck

5 months ago

Over the last 12 years, I've worked with 4 Private Equity funds at ZoomInfo. Here’s my 12-step playbook for working with PE Owners. In 2014, I took on my first PE investors (TA Associates and 22C Capital). In 2018, we brought on Carlyle and CPPIB. Private Equity is in the

63 47 917 372K 2K

@WhaleEverything @NoLimitGains @grok DONE

@ArrakisGlobal Who wrote the original research piece you cited?

Potentially one of the best developments to happen to our republic in a very long time! More power to the masses!

Anthony Pompliano 🌪 @APompliano

5 months ago

You are about to see hundreds of Nick Shirley imitators flood social media and YouTube. They will go all over the country trying to expose the insane fraud and wasteful spending. The incentive to capture millions of views is too great to be ignored by a generation of young

1K 6K 35K 1.7M 2K

@cjgustafson Everyone has to stop with the EV/sales multiples. Software is just an ordinary business, nothing special. The illusion that these businesses will just magically print money when they stop “investing” has been disproven. Come back to me when this trades at 20-25x PE

The structure of the deal and partnerships itself (tranche 2 and 3) and the fact that $HUT management @ashergenoot @bigsuey alluded to another party losing out on tranche 1 points to additional deals to come - perhaps multiple deals. Its very reasonable to assume that every deal $HUT announces adds another $30-40/share in equity value to the stock. Pretty easy to sketch out the stock doubling or tripling from here!

Capitalizing this one deal alone at 15x EV/EBITDA yields an equity value of $36/share. Add in another $7/share in BTC and cash and your at $43/share - basically today's stock price. What the market is missing and not pricing in is the next deal and the next deal after that. $HUT created $36/share in equity value using less than 10% of the power (MWs) that they have locked up.

$HUT signed a $7B revenue, 15 year agreement (with additional 15 year extensions) with Fluidstack (fully backed by $GOOGL AA+ credit rating) who will be renting capacity to @AnthropicAI This deal is one of the best infrastructure deals I have ever seen: 1/ $HUT set a new high watermark in terms of economics - NOI of $1.9M/MW (average BTC miner deals $.53) 2/ Triple net lease with all lease payments, utility payments, and maintenance costs fully backed by $GOOGL - one of the highest quality counterparties in the world 3/ 85% of construction costs will be financed by $GS and $JPM at a rate in the low 6% range (already locked in). This works out to an unlevered project yield of 19% (amazing!) and a levered equity yield of 66% (muy bueno!)