Longreads + Open Thread

Longreads

• Steven Levy in Wired has a wonderful profile of OpenAI. A consistent theme in the piece is belief: OpenAI deliberately recruited people who believed that artificial general intelligence was possible (and soon!) even though a) the idea that this was possible was not mainstream among researchers at the time ("Back in 2015, when we were recruiting, it was almost considered a career killer for an AI researcher to say that you took AGI seriously.") and b) OpenAI did not have a specific idea of how to get there. Faith in the outcome is a good way to handle disappointments along the way, which is particularly important for hard tech companies where there's a long wait for any evidence that something is working.
  
  (Between this piece on the faith-based AI lab and an article called "I Saw the Face of God in a Semiconductor Factory " a few months ago, Wired is becoming the premier source for news about applied theology in business and technology.)
• The other day I came across a shocking claim: there was once a software project that was finished, not just on time, but so early that the team decided to add a bunch of features before launch. That project was the video game Final Fantasy VI (for fellow members of the SNES generation who played the title: the Floating Continent was supposed to be the end of the game, and everything after that was added later). This oral history (Japanese; Google Translate link) describes the process. Key ingredients: a small team, at least by modern standards; a rival product that the team wanted to beat; reusing game assets so much of the extra content could consist of finding new ways to combine existing components; and sleeping in the office.
• A 2010 article investigates the story that Gillette used razors as a loss leader in order to sell blades and finding, based on contemporary catalog listings and Gillette's annual reports, that this was completely wrong. Gillette actually priced their razors at a premium, and won through products and marketing. Annoyingly, "razors-and-blades" is a catchphrase most people understand, so we're probably stuck with it—much in the way that someone buys a product whose usage requires some consumable good that can only be purchased from the original manufacturer at a ridiculous markup.
  
  Via Edwin Dorsey's Idea Brunch interview last week.
• This 2020 Institutional Investor profile of Dmitry Balyasny is a good look at how asset management has evolved over time. Balyasny started out at a day-trading firm before spinning off on his own, and there are some similarities between the modern multi-strategy model and the 90s-era day-trading business of finding a bunch of prop traders, firing most of them when they underperform, and then getting capital, exchange connectivity, and risk-management for the handful of winners. (Incidentally, his old trading firm, Schonfeld, now operates a multi-strategy hedge fund as well.) From the piece, a summary of what’s changed:: "In the day-trading days it was like running a casino. In the middle years it was like running a sports franchise. Now you are running a business. Expenses, costs become important,"
• This 2010 Malcolm Gladwell article on entrepreneurs and risk-taking is worth reading. Gladwell makes two core points: first, people who get very rich tend to do so in sudden spurts; even if their net worth theoretically compounds at a somewhat steady pace, the ability to compound at that pace can often be traced to a handful of decisions. (It's semi-reasonable to view the faster markups on early-stage investments compared to previous market cycles as a reflection of this pricing inefficiency.) The second and more practical point is that many of the deals that first catapult people into wealth are less risky than they look. He opens with a study of Ted Turner's decision to buy a money-losing TV station, noting that Turner already owned billboards that could be used to advertise it, and that programming for the station was cheap. It's pretty typical to evaluate high-risk investments by thinking about what the long-term reward could be—the TAM slide in the presentation, the bull case for the stock—but it might be a good mental exercise to hold the reward constant and start enumerating ways to get rid of risks.
• This week's Capital Gains goes in a slightly different direction than usual: how to use models from finance and economics to get to the crux of political questions. One point to highlight: the debate in many policies centers around the immediate distribution effects, but the impact of policies comes from long-term incentives those policies establish.

Open Thread

• Drop in any links or comments of interest to Diff readers.
• Following up on the Gillette story, as well as Ronald Coase’ essay on lighthouse economics (previously linked here), what are some other examples of a business model whose canonical exemplar did not actually follow it?

Diff Jobs

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

• A firm using machine learning to customize investments is looking for a data engineer with AWS experience. (NYC)
• A company building the new pension of the 21st century and building universal basic capital is looking for fullstack engineers with prior experience at a fintech startup. (NYC)
• A crypto proprietary trading firm is actively seeking systematic-oriented traders with crypto experience—ideally someone with experience across a variety of exchanges and tokens. (Remote)
• A new fintech startup wants to bring cross-border open banking to LATAM, and is looking for a founding engineer. (NYC)
• A concentrated crossover fund is looking for an intellectually curious data scientist with demonstrated mastery in analytics. Experience with alt data, web scraping, and financial modeling preferred. (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.