• AppLovin (APP) stands out as the strongest near-term opportunity: revenue grew 55%, free cash flow reached $1.4 billion, and management raised fourth-quarter guidance; consider buying into strength while recognizing competition and market-saturation risks.
• NVIDIA (NVDA) remains a long-term AI-demand holding, but data-center delays and uncertainty over spending growth could drive volatility; the discussed $250 year-end figure was an individual’s aspiration, not a firm target.
• Treat Nebius (NBIS) and CoreWeave (CRWV) as higher-risk growth investments: geographic reach and available capacity could help, but monitor project delivery, customer commitments, and financing needs before adding exposure.
• For less direct exposure to AI infrastructure, Netflix (NFLX) was cited as trading near a five-year valuation low, with roughly 20% projected EPS growth over the next few years; this was a speaker’s view, not a formal target.

AI infrastructure and data-center spending

• The central debate was whether data-center construction and AI capital expenditure (CapEx) could slow after election results and reports of permitting, power, and water constraints.
• One view was that restrictions could hurt chipmakers and other infrastructure suppliers, while allowing hyperscalers to spend less and potentially improve margins and free cash flow.
• The counterargument was that AI demand is global, many projects are outside the United States, and existing backlogs and commitments could keep investment elevated. The speakers also stressed that a slower rate of CapEx growth is different from CapEx actually declining.
• Citrini Research was cited as arguing that the risk to AI-infrastructure spending has shifted to the downside. The speakers disagreed over how significant that warning was and whether it was a change in outlook or a response to political headlines.
• Risks discussed included data-center moratoriums, permitting and power delays, expensive financing, and uncertainty about whether AI companies can turn their investment into sufficient revenue.

Takeaways

• Treat AI-infrastructure exposure as a range of outcomes, not a single forecast. A slowdown in spending growth could affect supplier valuations even if total spending remains high.
• Distinguish companies that directly depend on continued construction from diversified hyperscalers that can adjust spending, use other regions, or benefit from lower costs.
• The transcript offered no definitive answer on the scale or timing of any CapEx slowdown.

NVIDIA (NVDA)

• NVIDIA was down about 3% on the day but closed above $200. One speaker said a $250 year-end price was his hope, but reaching it in the remaining roughly 55 days seemed difficult; this was not presented as a firm target.
• The speakers connected the sell-off in semiconductors to uncertainty over data-center buildouts, as well as reports that Goldman Sachs was restricting some AI-tool use and Apple was reducing internal AI budgets.
• One speaker remained a long-term believer in AI demand and said he was not looking to sell or rotate out, while acknowledging volatility and the possibility of a lower close.

Takeaways

• The bullish case rests on continued AI demand and data-center investment. The key risk raised was that political or infrastructure constraints could slow new capacity and weigh on chip demand or valuations.
• The $250 figure was an individual’s aspiration, not a formal recommendation or assured outcome.

Micron (MU), Broadcom (AVGO), AMD and semiconductor ETFs

• Micron closed at $795, below $800 for the first time in about a month. Broadcom closed around $325, AMD around $488, and the SMH semiconductor ETF fell about 2.7%. The broader semiconductor group underperformed despite gains in the S&P 500 and Nasdaq.
• The speakers debated Micron’s low valuation against its strong expected earnings and demand for memory. The bullish argument was that AI workloads could keep memory demand strong and that Micron could generate substantial cash. The cautious argument was that memory remains cyclical, future supply could increase, and expected earnings growth may already be near a peak.
• The discussion also emphasized the possibility that the market could keep semiconductor valuations subdued even if companies continue to grow. AMD was described as having strong growth prospects, but the speakers disagreed on how much that translated into an attractive risk/reward at its then-current price.
• SMH received some buying near the close, but no specific investment recommendation was made.

Takeaways

• The debate is not simply whether AI demand exists; it is whether current earnings expectations and valuations adequately account for cyclicality, future supply, and slower growth.
• Investors considering semiconductor exposure should distinguish broad chip demand from individual companies’ sensitivity to memory pricing, data-center construction, and valuation changes.

Hyperscalers: Microsoft (MSFT), Alphabet (GOOGL), Amazon (AMZN), Apple (AAPL) and Meta (META)

• The major cloud and technology companies rose while semiconductors fell: Microsoft gained about 7%, Amazon about 6%, Alphabet about 3%, and Meta about 1.4%.
• The discussion suggested that the market may view lower CapEx as potentially beneficial to the hyperscalers’ margins and free cash flow, even if lower spending would hurt suppliers. The counterpoint was that constrained capacity could limit cloud growth and cause companies to lose market share.
• Alphabet’s announced $15 billion data-center project in Malaysia was cited as an example of investment outside the United States.
• Apple was reported to have reduced internal AI spending from $300 to $200 per employee. Speakers viewed the news as a negative signal for AI infrastructure sentiment, while noting that it did not mean data centers would stop operating.
• Microsoft reportedly lowered planned data-center spending after project delays involving third-party infrastructure providers. The report cited possible delays related to IREN and Nebius, though the speakers noted that some of the information might not be new. Nebius separately described delivering power for a customer site in New Jersey on a rapid timeline.

Takeaways

• The transcript highlighted a possible divergence: constraints may benefit hyperscalers through lower spending but harm their suppliers, while also creating a risk that limited capacity restricts cloud growth.
• Track actual project delays and company guidance rather than assuming that political headlines will either stop investment or have no effect.

Nebius (NBIS), CoreWeave (CRWV) and IREN

• Nebius and CoreWeave were discussed as cloud providers with available capacity. The speakers noted that some neocloud capacity may be sold through shorter-term deals, potentially allowing providers to benefit if spot prices rise.
• Nebius was described as having a geographically diversified data-center portfolio, including Europe and the Middle East. One speaker argued that this could position it well if U.S. construction becomes more difficult.
• Reports of Microsoft-related delays raised concerns about IREN and Nebius. The speakers disputed how much the reports indicated a real slowdown, pointing to Nebius’s own account of quickly securing approvals and making power available.
• A possible bullish scenario discussed for Nebius was a strong future ARR outlook supported by customer prepayments. The speakers cautioned that rapid growth could still require substantial financing, and a higher cost of capital could weigh on the shares.
• CoreWeave was also discussed as vulnerable to financing costs and investor concerns about continued capital raising.

Takeaways

• For these companies, growth announcements need to be assessed alongside financing needs, project delivery, customer prepayments, and the cost of capital.
• The transcript presented Nebius’s geographic mix and reported project execution as possible advantages, but did not establish that all reported delays were resolved.

Data-center power: Vertiv (VRT) and Bloom Energy (BE)

• Vertiv and Bloom Energy were cited as possible providers of on-site power solutions if data centers are required to supply their own electricity.
• Despite that potential opportunity, the speakers noted that related stocks were still falling amid uncertainty about data-center construction and possible state-level restrictions.

Takeaways

• The potential demand for on-site power may be an investment theme, but the discussion did not establish that it would offset delays or cancellations in new data-center projects.

Qualcomm (QCOM), Arm (ARM), Coherent (COHR) and optics suppliers

• Qualcomm reported revenue of $10.44 billion, above the $10.14 billion estimate, and a very small adjusted EPS beat of $2.17 versus $2.16 expected. It forecast first-quarter revenue of $10.4 billion to $11.2 billion and confirmed renewed agreements with Apple and Huawei. Its shares were down about 4–5% after hours.
• Qualcomm’s planned CapEx was reported at $2.5 billion, above the $1.9 billion expectation; the increase was cited as one reason for the mixed reaction.
• Arm beat on adjusted EPS and revenue guidance, but royalty revenue was lighter than expected. The company said its share of the server market at the largest hyperscalers had risen from 15% in 2024 to more than 50% in 2025, and it expected that to exceed 60% by the end of 2026. Its shares moved from an initial gain to around flat after hours.
• Coherent beat estimates by about 6% on EPS and 3% on revenue, with year-over-year EPS growth of 80% and revenue growth of 50%. Despite the results, its shares initially fell, reflecting weak sentiment toward optics and AI infrastructure.
• Lumentum, Applied Optoelectronics (AAOI), Marvell (MRVL) and Credo (CRDO) were mentioned as companies whose prospects could be affected by Coherent’s results. The speakers also discussed reported interest in Lumentum from other potential bidders after a Marvell-related acquisition announcement, but treated the reports as uncertain.
• The speakers questioned whether a proposed Marvell-Lumentum combination and its reported dilution reflected supply constraints or changing expectations for the optics market.

Takeaways

• Strong results did not guarantee a positive share-price reaction: sentiment about AI infrastructure and future capacity remained a major driver.
• For Qualcomm and Arm, monitor the mix of mobile, licensing, and data-center growth. For optics suppliers, monitor demand, project timing, and the terms of any proposed transactions.

AppLovin (APP)

• AppLovin reported revenue of $2.18 billion, up 55% year over year, and adjusted EBITDA of $1.84 billion, up 59%. It generated $1.4 billion in free cash flow and reported an 84% adjusted EBITDA margin.
• The company raised its fourth-quarter revenue outlook to $2.365 billion–$2.395 billion and adjusted EBITDA outlook to $1.995 billion–$2.025 billion. It also increased its share-repurchase authorization by $3.2 billion, bringing the total authorization to $3.6 billion.
• Management said a partial rollout of new advertising models had improved performance, with more rollout and optimization still ahead. It described consumer advertising as growing rapidly, and said gaming revenue had reaccelerated to nearly 50% year-over-year growth.
• Management said the business could compound at roughly 30% annually over the long term, while acknowledging that quarterly model improvements would not arrive on a predictable schedule. It also identified customer acquisition, creative tools, and connected TV (CTV) as areas of focus.
• The main concern discussed was whether AppLovin’s addressable market could become saturated or whether competitors could erode its performance. Management argued that advertisers were getting stronger returns and that the consumer business still had room to grow.
• The stock rose sharply after hours. One speaker called it a stock to “buy into strength,” based on the results and valuation; this was that speaker’s view, not a consensus recommendation.

Takeaways

• The quarter provided evidence of strong growth, cash generation, and management confidence. The key question remains whether new models and customer growth can sustain performance as the company gets larger.
• Investors should weigh the strong reported results against the transcript’s unresolved concerns about market saturation, competition, and the durability of growth.

MercadoLibre (MELI)

• MercadoLibre’s revenue was $10.75 billion, above the $10.7 billion estimate and up 45% year over year. EPS of $8.83 missed the $9.27 estimate by about 5%.
• The stock fell after hours. The speakers attributed the EPS shortfall and pressure on margins to heavy investment; management was said to be continuing that investment to gain market share.
• One speaker described the company as having grown revenue by more than 30% for five years in a row and compared its investment phase to Amazon’s earlier development. The discussion also noted that Mercado Pago could provide future operating leverage.

Takeaways

• The investment case presented was strong revenue growth and potential long-term margin improvement, balanced against near-term earnings pressure from reinvestment.
• The transcript offered no timeline for when margins might recover.

e.l.f. Beauty (ELF)

• e.l.f. reported EPS of $1.20 versus $0.60 expected and revenue of $520 million versus $473 million expected. Revenue was up 51% year over year and EPS was up 78%.
• The stock initially rose but later gave up much of its gain. The discussion noted that management attributed much of the EPS improvement to lower year-over-year tariffs, which made the headline earnings beat less clearly representative of underlying performance.
• Retailers were described as benefiting from the day’s market rotation, but no specific long-term recommendation was made.

Takeaways

• Separate underlying demand and operating performance from temporary or policy-related benefits. The tariff contribution was identified as an important qualification to the earnings beat.

Figma (FIG)

• Figma reported revenue of $391.4 million, up 43% year over year, compared with $376 million expected. EPS was $0.10, double the $0.05 estimate, though unchanged year over year.
• The stock fell about 8–9% after hours despite the beats.
• One speaker raised a competitive concern: an engineer had stopped using Figma for some design work and was using Codex instead. The transcript did not provide evidence that Figma was broadly losing customers, but the anecdote illustrated investor concern about AI tools changing software workflows.

Takeaways

• Strong quarterly results did not eliminate concerns about whether AI tools could change demand for Figma’s product.
• The anecdote is a risk to investigate, not proof of a broad competitive shift.

Snap (SNAP)

• Snap reported revenue of $1.72 billion, slightly below the $1.73 billion estimate, and EPS of $0.03 versus zero expected. Daily active users were 495 million, about 4 million below estimates.
• Its fourth-quarter revenue and EBITDA guidance were above expectations, but shares fell about 9–12% after hours.
• Snap said it would stop reporting daily active users and report monthly active users instead. The company had 977 million monthly active users. A speaker viewed the change as bearish because it reduces visibility into engagement.
• CEO Evan Spiegel said advertising revenue growth remained below the company’s ambitions.

Takeaways

• The key concerns raised were user engagement, advertising execution, and reduced disclosure around daily usage.
• Stronger forward guidance was a positive counterpoint, but the transcript’s overall view was cautious.

Dutch Bros (BROS) and Cava (CAVA)

• Dutch Bros reported EPS of $0.25, in line with expectations, and revenue of $549.8 million versus $548 million expected, with revenue up 30% year over year. Shares fell about 17% after hours.
• The decline was attributed to weaker-than-expected full-year guidance: adjusted EBITDA guidance was lowered to at least $380 million, versus a prior range of $385 million–$390 million.
• Cava reported EPS of $0.11 versus $0.12 expected and revenue of $356 million versus $358 million expected. It raised its full-year comparable-sales outlook to 6%–7% from 4.5%–6.5%, while slightly lowering the top end of its restaurant-level profit-margin range.
• Cava’s CFO reportedly said the margin change reflected reinvestment to support durable growth. Shares were down about 9%.

Takeaways

• For both restaurant stocks, guidance and margin expectations mattered more to the reaction than modest quarterly revenue beats.
• Cava’s improved comparable-sales outlook was positive, but its reinvestment plans could limit near-term margins.

Sunrun (RUN)

• Sunrun, described as a solar company, reported EPS of $0.56 versus $0.32 expected and revenue of $1 billion versus $805 million expected.
• Revenue was up 52% year over year and EPS was up 833%. The stock initially rose about 15% after hours, then gave back some of the gain.
• The speakers suggested the election outcome could improve investor interest in the company, but did not discuss specific policy changes or a price target.

Takeaways

• The reported quarter was strong relative to estimates, but the transcript did not establish whether the earnings growth was sustainable or how much it depended on policy or one-off factors.

Fastly (FSLY)

• Fastly beat EPS estimates, reporting $0.17 versus $0.12, and revenue of $190 million versus $187 million. Revenue was up 20% year over year and EPS up 143%.
• Despite the results, shares fell about 10–14% after hours.
• The discussion linked weakness in content-delivery networks and related infrastructure stocks to concern that data-center and AI spending may be less certain. Fastly was also discussed as potentially relevant to the growth of AI agents and digital infrastructure.

Takeaways

• The market reaction suggested that investors were focused on the outlook for AI-related infrastructure demand, not just the reported quarter.
• The transcript characterized the sell-off as potentially excessive, but did not provide a specific valuation or price target.

Palantir (PLTR)

• Palantir was described as potentially better positioned if investors shift toward companies with durable revenue growth, strong margins, and less direct dependence on data-center CapEx.
• It was down about 2% on the day, less than some infrastructure-linked stocks. A speaker said the company should be “okay” if it continues to deliver earnings, while acknowledging political-sentiment concerns tied to its associations.
• The broader argument was that future investors may favor companies with sustainable earnings over companies dependent on every component of GPU and data-center spending.

Takeaways

• The transcript presented Palantir as a possible relative beneficiary of a shift toward profitable software and away from capital-intensive AI suppliers.
• That view depends on Palantir continuing to deliver the earnings and growth that speakers cited; no price target was given.

Reddit (RDDT)

• Reddit came up in a discussion of whether a sharp post-earnings fall could be a buying opportunity or a sign of a lasting problem.
• The main risk raised was reliance on Google search for user discovery. Speakers questioned whether Reddit could continue growing users if that channel weakens.
• Potential positives mentioned included continued user monetization, rising average revenue per user, and possible AI-data licensing deals, though no specific deal was cited.
• One speaker said he had not sold his shares but was not yet ready to add, pending further review of the results and the search-related concerns.

Takeaways

• The key issue is whether search-driven discovery is essential to Reddit’s future growth or whether the company can acquire users through other channels.
• The speakers treated the sell-off as uncertain rather than clearly calling it a buying opportunity.

Fintech and consumer-growth names: SoFi (SOFI), Dave (DAVE), Chime and Grab (GRAB)

• SoFi was described as a strong business with a long-term growth opportunity, despite a weak stock performance that one speaker characterized as a significant opportunity cost that year. Management’s outlook was cited as evidence of confidence in the company’s growth.
• Dave reported revenue of $198 million versus $192 million expected, up 31% year over year. EPS was $4.39 versus $4.33 expected, but the shares fell about 10%. Adjusted EBITDA was slightly below expectations.
• Chime was described as beating on revenue and EPS, raising guidance, and reporting more members than expected. Its shares rose about 6%. The speakers compared it broadly with other fintech and digital financial-services companies.
• Grab was described by one speaker as executing well and continuing to establish itself as a real, growing business. The speaker acknowledged that the stock had fallen substantially and said there was still a long way to go.

Takeaways

• The discussion favored evaluating these firms on execution, customer growth, and guidance rather than relying solely on short-term share-price moves.
• The transcript did not provide specific price targets or detailed risk analysis for SoFi or Grab.

Netflix (NFLX) and Uber (UBER)

• Netflix and Uber were mentioned as potential alternatives to companies more directly tied to AI CapEx.
• A speaker described Netflix as trading near a five-year valuation low and projected roughly 20% EPS growth over the next few years. The same speaker said the business was strong despite not particularly liking its content.
• Uber was listed as a possible company that could benefit from long-term growth without depending directly on AI infrastructure investment. No company-specific figures were discussed.

Takeaways

• The speakers’ broader preference for 2027 was for diversified, established businesses that could grow without requiring continued acceleration in AI-infrastructure spending.
• These were examples in a portfolio discussion, not explicit buy recommendations.

Other companies discussed

• Duolingo (DUOL) was described as the prior day’s biggest outperformer, up 34%; later, a speaker noted it was up about 25% on the day and praised its long-term thesis. No new earnings figures were discussed.
• Zeta Global (ZETA) was described by one speaker as a successful holding and a company he would be more interested in buying around $32–$34. This was his personal view, not a consensus price target.
• Redwire (RDW) reported revenue of $109 million versus $126 million expected and negative free cash flow of $67 million versus negative $21 million expected. EPS beat estimates, and its full-year revenue outlook was above estimates. Shares fell about 5% after hours.
• Joby Aviation (JOBY) reported revenue of $55 million versus $40 million expected and beat on EPS. The speakers noted that the figures were small; no further investment view was given.
• Centrus Energy (LEU) reported revenue of $55 million, slightly below expectations, and EPS of negative $0.46 versus negative $0.39 expected. Its revenue outlook was above estimates, and its shares were roughly flat after hours.
• FICO was said to have beaten on EPS and revenue, with year-over-year EPS growth of 26% and revenue growth of 27%. Its shares nevertheless fell about 12% after hours. The discussion cited the absence of 2027 guidance as a possible reason.
• AMC Entertainment (AMC) beat revenue and EPS estimates, with revenue up 32% year over year. One speaker expressed little interest in the stock, describing it as a former meme-stock trade.
• AMC, Expedia, HubSpot, Motorola, Beyond Meat, Wolfspeed, Sunrun, Kava/Cava, Figma, Fastly, and other names were also mentioned in the earnings rundown, but the transcript did not provide enough company-specific investment analysis for a fuller conclusion.