Leandro

Just a thought but what if Amazon launches an agent that can look outside the Amazon marketplace and starts to roll out “Buy With Prime” seriously? You’d get organic traffic from prime members and the Amazon experience (Obv impossible to know how everything will evolve) $AMZN

Think a lot of people are underestimating Amazon’s role in all of this. Of course it’s still very early in agentic search, but there’s much more to Amazon’s value prop than low prices and a lot of it can only be achieved with its ecommerce infrastructure $AMZN

Don't think I entirely agree here. Meta's ads are for sure a byproduct of entertainment but what would avoid someone from seeing something on Meta and taking that intent out of the platform (maybe to an entirely different company that the one advertising on Meta) through an agent? As for Amazon, we don't really have data on what happens in the discoverability realm and what happens after a purchase intent has already been typed in which is likely important to judge the impact of agentic search

Maybe wrong here and we don't have data, but seems like a good chunk of Amazon's marketplace ads happen post-purchase intent/search, so not really extremely exposed to discoverability Also, why wouldn't a personal agent be bad for Meta's ad business in the future? (assuming agents get "commoditized"). Seems like there is more discoverability exposure in Meta ads that could be substituted by agentic search Just thinking out loud here $META $AMZN

I liked how Jensen Huang framed the paradigm shift that AI brings to the technology industry: "When you touch something on the phone, it goes and retrieves the piece of information. It was all pre-recorded. If you want to ask anything and know everything, you want to actually know it, you do not want to find it, then you have to generate the answer. You cannot reasonable retrieve that. A new layer of computers that is continuously generating answers based on all the queries that are coming at it has to get built. There's no longer a bunch of storage, it's a bunch of computers." One interesting thing here is that we seem to be in the early innings of this paradigm, as the remaining layers seem to be much more compute intensive: - Layer 1: humans asking LLMs - Layer 2: agents asking each other in an automated manner (we are here but still lightly penetrated) - Layer 3: physical AI continuously iterating on what it is seeing in the real world Jensen argues that layer 2 is 10x the compute of layer 1, with layer 3 being 10x the opportunity of Layer 2. Of course the timeline is unknown and bottlenecks may appear, but it seems reasonable that we'll get to layer 3 eventually and that a good chunk of the new infrastructure still needs to get built. Reasonable doubt: will the builders get their fair share of economic profit or will it accrue to consumers?

@BornInvestor I am also long NVDA and can see how this is good for the picks and shovels, just have a tough time seeing how its good for the labs (which evidently feeds into the supply chain if they can't make their margins)

The $IGV today after strong $CRM, $VEEV and $NVDA earnings https://t.co/MYKG8ppEEQ

My good friend @AboveAvgOdds wrote the following in his most recent investor letter: "Nearly the entire first half's returns showed up in roughly tend week off the March lows, and it went to whoever had simply waited. None of this is a 2026 phenomenon. It's the oldest asymmetry in the business, and it's why waiting for clarity can be the most expensive trade there is: Clarity and the rally are the same event." I couldn't agree more with Ryan, and a good chunk of my investment philosophy has been molded by something similar. The quote below (courtesy of my friend's @JRogrow interview to Jerome Dodson) had a profound impact on my investment philosophy: "Many of our biggest winners have been companies that operate in cyclical industries with secular growth drivers. When their business cycle turns down, investors become overly pessimistic and extrapolate the current negative conditions. They forget the cycle will eventually turn, throw in the towel on the secular growth drivers, and engage in panic selling, pushing the stock to bargain-basement levels. But eventually the cycle turns, and the stock soars higher. It’s difficult to have the courage to buy when everyone else is selling, and this has been an important part of our success." Being able to wait for clarity to arrive or acknowledging that one can't precisely time inflections is a source of alpha. Even though I used to "hate" cyclicality, I actually currently think that it's the scenario where most of these situations present themselves (so long as there's secularity to the business). I've seen this in a couple of businesses that have made it to the portfolio over the past couple of years like... $ASML 25% CAGR (since inclusion) $DE 16% CAGR $KEYS 55% CAGR $MEDP 52% CAGR All of these had one thing in common: swarms of investors were waiting for clarity to arrive, thinking they could time the inflection to perfection. The risks are always screaming at you at the lows, that's why investing is psychologically challenging.

Nvidia is trading at less than 11x Wall Street’s financing and you are bearish? $NVDA https://t.co/2wNVBgvx4l

The market thinks hyperscaler ROIC will be subpar while bidding up the entire data-center supply chain. Both can't stay true and you can’t sell the picks while doubting the miners. $AMZN Q2 may have marked the inflection and I explain why in my most recent (free) post! https://t.co/7Ty0yhyWQh