Longreads + Open Thread

Longreads

• Sarabet Chang Yuye has a fun and detailed review of A Brief History of Intelligence, which answers some helpful questions on how life could go from simple nervous systems (if you recognize something as food, go towards it; if you recognize it as danger, scoot away) to a species capable of, say, writing long blog posts on how intelligence arose. Because we have access to random messy snapshots of evolutionary history, and because we have better information on what ancient life was shaped like rather than what it acted like, this is rife with opportunities for just-so stories. But if that's what these are, they're all pretty good: it's particularly helpful to see emotion as emerging from the 2x2 matrix of energy expenditure level and opportunity/danger. There's also a great bit towards the end that helps to explain two confusing stylized facts: humans don't seem to have much speech-specific hardware that other primates lack, but we're better at language, and the theory of language evolving to support group coordination requires a lot of activation energy. But if language actually started out one-on-one, between mothers and infants, that neatly solves both problems. (Via Snippet.finance.)
• Tyler Cowen interviews George Selgin, who has a number of contrarian takes on the Great Depression and policy. Selgin's a fun thinker, because, as he notes in the interview, he's rabidly libertarian on basically one issue—that you don't strictly need the government to run the monetary system. Debatable for sure, but it means evaluating a time period like the Great Depression/New Deal era, with its wild credit fluctuations, substantial banking regulation overhauls, etc. from a different angle.
• And, with excellent timing: Stein Berre and Asani Sarkar at the Liberty Street Economics blog look at how the Dutch had a reserve currency without US/UK-style central banking. We've written a bit before about how entrepôts lead to financial markets: if there's a city that specializes in grading, storing, and reshipping commodities, it's a great place to bet on commodities. When moving money around means putting a box of coins or bullion on a ship, that money will naturally gravitate to where it has the most opportunities to be deployed. A country that has trade relationships everywhere is one that's expensive for anyone to invade, and has very strong incentives not to expropriate, so it's a low-cost borrower. And all those features have feedback loops. But that also leads to a very levered financial system, and high asset prices. So it can't last forever.
• Derek Thompson on how social media, podcasts, and AI are all converging on television. It's hard to articulate exactly what makes TV so compelling (though The Diff has tried), but one reason might be that it's the medium where passively consuming ads is easiest, so it monetizes best, so it's the kind of content that rewards investments in quality (or at least in making what viewers compulsively watch).
• Rob Horning also has thoughts on generative video. Specifically: who actually wants to watch this stuff? That remains a mystery to me, too; there's some fun novelty, but it's hard to imagine sitting down and watching two hours of generative AI clips instead of watching an actual movie. But this is the kind of question that you can always ask early in a medium's existence. Nobody knew what video, radio, or even the written word were uniquely good for at first, either. (In fact, our earliest written narratives are transcripts of older oral traditions, while many of the earliest samples of writing we have are business records. Maybe to an ancient Sumerian, the idea of writing a novel would sound as counterintuitive as creating a choose-your-own adventure story entirely in Excel.)
• This week in Capital Gains: how do frauds happen? The ones that are born as frauds tend to die early, and the one big exception, Madoff, had a non-fraud operating business attached.
• One question from a ReadHaus user: when is it possible for a business to sell data? One meta point here is that in many cases, the optimal setup is being able to use the data—if you harvest information and use it entirely within your own business, you basically have perfect price discrimination with respect to your data monetization. And that means you can afford to collect a lot more of it.

Books

1929: Inside the Greatest Crash in Wall Street History--and How It Shattered a Nation: If you want to, you can draw plenty of parallels between the US's economic and financial situation in the late 1920s and today. You have the backdrop of rapid development of some new technology (electricity—which, like AI, had consumer and business applications and changed behavior in very different ways for each), rising geopolitical tensions which sometimes take the form of protectionism, lots of retail speculation in the stock market, companies participating in pumping up their own shares, and way too much leverage.

But that time was different. Not in a way that we can't learn from, but in a way that requires some judicious adjustments. Markets moved at a different pace in the 1920s, with a long lag between submitting orders and getting them executed, and sometimes a multi-hour delay between when trades happen and when the ticker tape finally prints them. The trading scenes in the book are a reminder that we should never take modern markets' latency for granted—even if shaving microseconds off order speeds doesn't seem like a socially-valuable activity, it's almost hard to see from a modern standpoint how markets could have functioned at all when things were so slow and uncertain. If you don't know where things are trading, but the last you've heard is that they dropped, and you want to sell, do you place a limit order to avoid getting ripped off? Or a market order so at least you know the trade executes? And, if you do that, what's it going to be like waiting an unknown length of time to see if your order actually made it through and, if so, what the price was?

The market was also a lot more of a literal, physical space; people in the book spend a lot fo time bounding uptown and downtown for meetings (and sometimes, conscious of appearances, to parties). That adds a lot to the drama: there's just something more evocative about an important trader striding onto the floor of the NYSE and lobbing a monster order for US Steel to stop a collapse, compared to the modern equivalent, of someone quietly tweaking the parameters of a VWAP order.

One of the big lessons from the book is that investor populations matter. The Fed was worried about speculation well before the peak, and Hoover was also not a big fan of speculators (he'd made his fortune in mining, on the operating rather than financial side). But the traders who would have responded to that were not the main price-setters: there was a new population of retail buyers, and a set of apex predators who made their fortunes exploiting these naive investors. Neither was really in a position to read into the Fed's signals the way a multi-decade finance veteran would have. The other important population was banks, specifically banks who knew that they'd risk a run if their stock dropped too quickly, and that this made it important for them to support their stock price. The opening scene in the book is First National Bank's CEO finding out that the bank blew a giant hole in their balance sheet by buying back stock aggressively to support the price. (As another book about the Depression notes, banks traded at absolutely crazy multiples at the peak.) All it takes to have the setup for a crash is for everyone to slightly overestimate the degree to which they know what they're doing.

Open Thread

• Drop in any links or comments of interest to Diff readers.
• The 1929 comparison is too easy, but are there other market periods that are worth thinking about more today? The last time we had a big capex buildout with lots of complementary companies might have been in the late 19th century: railroads needed steel (and transported it), and steel also helped raise the density of big cities—increasing the value of railroad transportation even more! 

Diff Jobs

Companies in the Diff network are actively looking for talent. See a sampling of current open roles below:

• A hyper-growth startup that’s turning the fastest growing unicorns’ sales and marketing data into revenue (driven $XXXM incremental customer revenue the last year alone) is looking for a senior/staff-level software engineer with a track record of building large, performant distributed systems and owning customer delivery at high velocity. Experience with AI agents, orchestration frameworks, and contributing to open source AI a plus. (NYC)
• A transformative company that’s bringing AI-powered, personalized education to a billion+ students is looking for elite, AI-native generalists to build and scale the operational systems that will enable 100 schools next year and a 1000 schools the year after that. If you want to design and deploy AI-first operational systems that eliminate manual effort, compress complexity, and drive scalable execution, please reach out. Experience in product, operational, or commercially-oriented roles in the software industry preferred. (Remote)
• Ex-Bridgewater, Worldcoin founders using LLMs to generate investment signals, systematize fundamental analysis, and power the superintelligence for investing are looking for machine learning and full-stack software engineers (Typescript/React + Python) who want to build highly-scalable infrastructure that enables previously impossible machine learning results. Experience with large scale data pipelines, applied machine learning, etc. preferred. If you’re a sharp generalist with strong technical skills, please reach out. (Remote)
• A startup is automating the highest tier of scientific evidence and building the HuggingFace for humans + machines to read/write scientific research to. They’re hiring engineers and academics to help index the world’s scientific corpus, design interfaces at the right level of abstraction for users to verify results, and launch new initiatives to grow into academia and the pharma industry. A background in systematic reviews or medicine/biology is a plus, along with a strong interest in LLMs, EU4, Factorio, and the humanities. (Toronto, Remote)
• Fast-growing, General Catalyst backed startup building the platform and primitives that power business transformation, starting with an AI-native ERP, is looking for expert generalists to identify critical directives, parachute into the part of the business that needs help and drive results with scalable processes. If you have exceptional judgement across contexts, a taste for high leverage problems and people, and the agency to drive solutions to completion, this is for you. (SF)

Even if you don't see an exact match for your skills and interests right now, we're happy to talk early so we can let you know if a good opportunity comes up.

If you’re at a company that's looking for talent, we should talk! Diff Jobs works with companies across fintech, hard tech, consumer software, enterprise software, and other areas—any company where finding unusually effective people is a top priority.