FundamentalAnalysis

87 posts

FundamentalAnalysis

FundamentalAnalysis

@FundyAnalysis

7 years in software IB and private equity investing & 5 years (so far) in L/S equity investing in TMT. I like to opine & share some personal wisdom I’ve learned

San Francisco, CA Katılım Mart 2025
102 Takip Edilen91 Takipçiler
FundamentalAnalysis
FundamentalAnalysis@FundyAnalysis·
$DDOG faces significant revenue concentration and churn risk as OpenAI, its largest customer at ~$200M+ revenue, migrates log, metrics, and trace workloads to self-managed Clickhouse/Grafana and Chronosphere stacks, implying a potential ~$100M revenue headwind and ~30% ARR decline by YE’27
English
0
0
0
1.8K
FundamentalAnalysis
FundamentalAnalysis@FundyAnalysis·
@dylan522p The Sightline number is announced GW not under-construction GW. Most ‘canceled’ projects never had an interconnection queue slot or PPA. Hyperscaler ’26 capex unchanged at $650B+. The delay is conversion lag. Also 1/2 of announced builds slip most were never real.
English
0
0
9
3.1K
Negligible Capital
Negligible Capital@negligible_cap·
*ANTHROPIC CONSIDERING FUNDING OFFERS AT OVER $900 BILLION VALUE Anthropic considering raising at over a $900 billion valuation, potentially surpassing OpenAI officially. The considerations are at an early stage and the company has yet to accept any offers according to sources. Bloomberg also notes that Anthropic had previously resisted several inbound proposals from investors for a new round at a valuation of $800 billion or more. $SKM and $ZM moving up after hours
Negligible Capital tweet media
English
3
9
62
10.3K
FundamentalAnalysis
FundamentalAnalysis@FundyAnalysis·
@bubbleboi 288 E-cores is a Sierra Forest profile not a Grace replacement. Would wonder if NVLink-C2C survives off an Intel-codesigned part. Without coherent fabric integration you’ve lost the reason Grace existed in the first place.
English
1
0
0
1.6K
bubble boi
bubble boi@bubbleboi·
My biggest out there Intel predictions. 1. Nvidia will replace their Grace CPUs with a co-designed X86 CPU that will have 288 cores or more of mainly E-cores. 2. Nvidia will use Intel advanced packaging - EMIB-T & Foveros for Feynman which will include stacked SRAM dies on 18/14a as well as I/O dies. 3. 14A 1.0 PDK will have a sizable advantage over TSMC’s 2nm and A14 node but I expect A14 to be delayed past 2028 giving intel a year head start.
English
31
30
573
82.8K
FundamentalAnalysis
FundamentalAnalysis@FundyAnalysis·
The setup going into 2025 was that Gemini and ChatGPT would structurally impair Search monetization. Consensus modeled mid-single digit Search growth decelerating into AI cannibalization. Q1 prints Search +19% Y/Y! Search revenue lives in commercial intent queries like “mortgage rates,” “lawyer near me.” AI Overviews cannibalize informational queries that barely monetized anyway. High CPC queries still surface ads inside AIO. Bears confused query mix with query count. Also, the multiple re-rate is not b/c Search is great but because the stack is finally vertically integrated and working. TPUs → Gemini → Cloud → Workspace → Search → YouTube. Every layer compounds the others. Apple is the only comp, and Apple doesn’t have hyperscale infra or autonomous vehicles. $GOOG is structural winner
English
0
0
0
428
Boring_Business
Boring_Business@BoringBiz_·
How is Google search still growing 19% year over year with nearly monopoly like market share in the category? Legitimately might be the greatest business ever created in the history of capitalism
Sundar Pichai@sundarpichai

Q1 earnings are in: 2026 is off to a terrific start. Our AI investments and full stack approach are lighting up every part of the business: Search queries are at an all-time high with AI continuing to drive usage. Google Cloud revenue grew 63%, Gemini models have incredible momentum, and it was our strongest quarter ever for consumer AI subs, driven by @GeminiApp. Thanks to our partners + employees around the world. Much more to share on our earnings call in 20 minutes… and at Google I/O in 20 days!

English
158
172
4.4K
336K
FundamentalAnalysis
FundamentalAnalysis@FundyAnalysis·
@citrini The Sightline number is announced GW not under-construction GW. Most ‘canceled’ projects never had an interconnection queue slot or PPA. Hyperscaler ’26 capex unchanged at $650B+. The delay is conversion lag. Also 1/2 of announced builds slip most were never real.
English
0
0
4
2.8K
FundamentalAnalysis
FundamentalAnalysis@FundyAnalysis·
@wallstengine 2.45 GW is bigger than $BE entire 25-year cumulative installed base (~1.5 GW through 2024). FY25 run-rate is ~350-400 MW of acceptances. Even at 2x current capacity, this is a 4-5 yr deployment. Caped and nwc need to scale..
English
0
0
2
812
Wall St Engine
Wall St Engine@wallstengine·
Oracle said it will now power its Project Jupiter campus in New Mexico with up to 2.45 GW of Bloom Energy $BE fuel cells instead of gas turbines and diesel generators. The switch cuts NOx emissions by about 92% and keeps the AI data center on a single microgrid campus.
Wall St Engine tweet media
English
27
116
968
363.1K
FundamentalAnalysis
FundamentalAnalysis@FundyAnalysis·
@sidtriv Sounds like massive waste of time. Unless you’re in growth or venture land
English
0
0
0
53
Sid Trivedi
Sid Trivedi@sidtriv·
Diligence in 2026 is wild. My friends in PE are now spending the weekend before IC trying to rebuild the company they're acquiring in Claude Code. If the clone works, the deal dies. Cheapest moat test in human history.
English
80
90
2.9K
485.1K
FundamentalAnalysis
FundamentalAnalysis@FundyAnalysis·
The investable stack this opens up, ranked by how tight the bottleneck is and how directly DPA dollars flow: (1) grain-oriented electrical steel - one US producer, every transformer needs it, no substitute; (2) large power transformers and GSUs -> 3+ year backlogs, (3) HV switchgear, breakers, and protective relays -> explicitly named in the determination, dominated by a handful of OEMs (4) HVDC converter stations and FACTS equipment -> the long-haul transmission spend tied to data center load (5) overhead conductor and ACCC/ACSS reconductoring -> the cheapest way to add transfer capacity without new ROW. The non-obvious tier sits one layer upstream: HV bushings, on-load tap changers, amorphous metal ribbon, copper magnet wire; components with effectively zero US capacity and no near-term substitute.
English
0
0
1
423
Emini tic
Emini tic@TicTocTick·
Grid is national security. Just in from the White House. Expect monster gap up in utilities next week.
Emini tic tweet media
English
59
152
1.5K
233.8K
FundamentalAnalysis
FundamentalAnalysis@FundyAnalysis·
The binding constraint isn't transformer assembly but itwo layers upstream. Grain-oriented electrical steel (GOES) is effectively a domestic monopoly with one US producer; without a second melt-and-pour line, every transformer subsidy just bids up core steel. Same story in HV bushings, on-load tap changers, and amorphous metal ribbon all import-dependent, all multi-year lead times. The investable shelf, in order of how tight the bottleneck is: (1) electrical steel and specialty alloys, (2) HV power transformers and large GSU units, (3) HVDC converter stations and FACTS gear (the AI/data-center load bridge), (4) MV/LV switchgear and protective relays, (5) overhead conductor and ACCC/ACSS reconductoring. the alpha is in the consumables and components nobody can substitute around. Federal money just makes the pricing power more durable for whoever owns the chokepoint.
English
0
0
0
144
Anthony Pompliano 🌪
Anthony Pompliano 🌪@APompliano·
Wow. The White House just announced that grid infrastructure is essential to national defense. This includes transformers, transmission lines and conductors, substations, and high-voltage circuit breakers. Companies working to electrify America will have a big tailwind.
Anthony Pompliano 🌪 tweet media
English
352
985
7.8K
1.1M
FundamentalAnalysis
FundamentalAnalysis@FundyAnalysis·
The 2018-22 software LBO model assumed ~95% gross retention and 110%+ NRR were structural. If AI compresses NRR to 95-100% and gross retention to 88-92% on mature SaaS, the debt capacity at any given multiple collapses not because cash flows go negative, but because the volatility of forward cash flows is no longer dischargeable. That’s why the dividend recap window is shut and continuation vehicles are flowing -> LP version of extending & pretending
English
0
0
31
4.4K
Boring_Business
Boring_Business@BoringBiz_·
Probably the best take I have heard on why there is so much fear around software exposure in private equity and credit because of AI From David Sacks "Historically we only had two good exits for software businesses. One was IPO. The other was M&A. Then these big private equity shops came along and gave us a third potential exit. You would sell to them and then they would raise the capital based on one third equity and two third debt. So it was debt financed buyouts. It is something that has been around in the non technology part of the economy for a long time but was a relatively new entrant in the world of technology. And the reason for that is if you have debt financed a purchase, you need to have very stable cash flows. Because if you miss and you cant pay your interest on the debt, you will lose all your equity because the debt holders will foreclose. It was belief for a long time that software did have predictable cash flows, at least for the mature businesses."
English
47
116
2.1K
400.7K
FundamentalAnalysis
FundamentalAnalysis@FundyAnalysis·
The main and arguably only reason $INTC is doing well right now is RL. Most reinforcement learning workloads run on CPUs. AI labs are spinning up millions of parallel environment containers. The bottleneck isn’t keeping a GPU busy all day anymore but it’s “I need millions of CPUs for the next 6 hours while my GPUs are tied up training the model.” GPU wait time is extremely expensive, so labs distribute rollouts across environments in parallel. On top of that, the new generation of long-horizon, multi-turn tasks can run for hours each where you can’t have GPUs sitting idle waiting on any single one. This is the same architecture being used at runtime for inference, code verification, and similar workloads. As we move into agentic AI, environments become a core part of both learning and inference even for enterprise workloads. The loop is -> agents are handed problems, they write code and call tools, and that code/tool execution happens inside a VM container. Labs spin up millions of these to verify the 100M+ tasks they’re running through. Each task can take anywhere from a few minutes to several hours, and you can’t let the GPU idle for that duration. So labs fan out hundreds of thousands of tasks in parallel across millions of these VMs.
English
0
0
0
192
FundamentalAnalysis
FundamentalAnalysis@FundyAnalysis·
My bet is this deal doesn’t close at $60B. Elon ends up buying the company for $10B instead. Cursor is at a $2.5B run rate, targeting $7B by year-end, on a negative 23% gross margin. Claude Code gives you $2,000 in credits for a $200/month seat and that’s the cost structure Cursor is eating. So if they hit $7B in revenue, Elon is writing $10B in checks to Anthropic. And Elon is not exactly known for paying his vendors on time. The only way this pencils is if Cursor migrates off Claude and onto xAI model. But problem is, xAI is probably the furthest behind on enterprise-grade and safety-hardened use cases. It’s great for consumer stuff on X but that’s a different bar. One angle that might actually interest Elon is Cursor was building its own model. So this could be a buy-for-the-model play, not a buy-for-the-coding-product play.
English
0
0
0
171
Sheel Mohnot
Sheel Mohnot@pitdesi·
Cursor at -23% gross margin in January $2.7B annualized revenue, up 14x YoY, expects $7B eoy but Claude code caught up fast, hard for cursor to raise xAI only at $3.2B 2025 rev, mostly twitter, really wants Cursors revenue 🍿 to see how it shakes out theinformation.com/articles/behin…
Sheel Mohnot tweet media
English
19
20
346
94.3K
FundamentalAnalysis
FundamentalAnalysis@FundyAnalysis·
@dylan522p Tokens to build it: $50K. Running it for 3 years (hosting, on-call, security, integration maintenance, feature parity with Slack’s roadmap): infinite time suck
English
0
0
3
943
Jason Kam
Jason Kam@MapleLeafCap·
Was chatting with one of the smartest guys I know in public equities this week and talked about SaaS stocks, his answer still sticks with me. Me: You spending any time in SaaS land at all? Him: You're not talking to somebody who woke up a loser. SaaS is gonna play defense for the next decade, that's a loser position, I don't back losers, I don't short exponential rise of intelligence that will iron out all friction between atom and bits. He obviously watches the Jensen interview.
English
30
18
401
100.6K
FundamentalAnalysis
FundamentalAnalysis@FundyAnalysis·
Two separate financings, not a refi loop. The $40B March bridge is already being termed out via last week’s $3.6B bond deal (10Y priced at 8.5%). This $10B margin loan is incremental leverage, ~425bps over SOFR = ~7.88%. I think real risk is simpler where concentrated private co collateral + levered pyramid on a single mark. If next OAI round prints below the last tender, margin call mechanics get ugly. Credit market already knows with CDS at ~360bps, near 1Y highs.
English
2
0
1
280
Serenity
Serenity@aleabitoreddit·
I'm sorry, but how does this work? > Softbank takes out $40B bridge loan to invest in OpenAi > OpenAi paper valuation goes up > Softbank gets margin loan off OpenAI's new valuation... maybe to pay back original loan? I'm getting PTSD from the FTX equity/FTT token but with OpenAI equity.
Serenity tweet media
English
69
42
788
147.2K
FundamentalAnalysis
FundamentalAnalysis@FundyAnalysis·
Yep, short-thesis on US solar/energy was more of a cost of capital bet. Chinese polysilicon made domestic manufacturing uneconomic at any reasonable WACC. The bear case on US manufacturing assumed capital would always flow to the lowest-cost node. B/w CHIPS, IRA, export controls, and hyperscaler onshoring, domestic names now have pricing power their predecessors never had. AMSC, NVTS, and the broader cohort are trading on a regime shift where American-made went from cost center to margin driver. Rare moment to be long the home team and the fundamentals 🇺🇸 ⚙️
English
4
0
4
2K
Serenity
Serenity@aleabitoreddit·
Just a shower thought... it’s extremely patriotic and positive sum to be investing in American National Security and supply chains like $AMSC or $NVTS. For the longest time hedge funds shorted domestic American sectors like energy + Solar to the point of bankruptcy... By betting with China on their state subsidies. and many years later, as a result: we have little excess energy capacity needed for AI compared to China. By investing alongside the White House and $NVDA to securing US supply chains with up and coming American companies… Stocka goes brrr? US companies go brrr? And America goes brrr? (Disclosure: I’m American, and heavily biased toward US)
Serenity tweet mediaSerenity tweet media
Serenity@aleabitoreddit

Yes. The President invoked the "Defense Production Act" 2 days ago. To expand on domestic infra. Implicit beneficiaries were: Transmission & Advanced Conductors: $AMSC - HTS Wires (advanced conductors), power electronics $PLPC - advanced conductors $ATKR - transmission components $VMI - grid components Power Electronics $AEIS - power control electronics $VICR - power delivery Capacitor Banks $VSH - power capacitors and capacitor banks Substations & High-Voltage Circuit Breakers $POWL - substations/switchgear $AZZ - substations Transformers $HPS.A - My favorite for transformers $SPXC - power transformers Electrical Core Steel $CLF - Only producer of electrical steel in America? This included: -Transformers - transmission components - advanced conductors - power electronics - substations - high-voltage circuit breakers - protective relays, capacitor banks - electrical core steel TLDR: The executive action declared a national energy emergency regarding the domestic supply chain for grid infrastructure. Authorized federal funding, purchase commitments, and expedited actions to rapidly expand the manufacturing. Lot of beneficiaries, I made a mini ETF of with $AMSC, $CLF, $PLPC, and then $HPS.A; dont have any of the others. There's probably going to be a bunch of DOE contracts in the next 3-6 months like " DoE Awards ___ to $POWL " off this act is my guess.

English
98
73
1.2K
307.8K
FundamentalAnalysis
FundamentalAnalysis@FundyAnalysis·
@Midnight_Captl Google Cloud’s 4th gen CDU or the boiler room of a 1970s NYC apartment building? Genuine question
English
0
0
2
104

Keşfet

@HyperTechInvest @dylan522p @negligible_cap @bubbleboi @citrini @wallstengine @sidtriv @elonmusk