How to get Inventors and Entrepreneurs to work together sustainably

A failure mode of tech-forward people and organizations is to barrel forward developing their technology, while taking a wait-and-see approach to what existing problems it could solve. Conversely, a lot of business-minded folks have a lot of great ideas for solving problems, but a limited sense of how to generate the necessary technology. It’s like there’s one fiefdom building nothing but hammers, and another making nothing but nails, and they hardly know how to talk to each other. This is a blueprint for vaporware companies and orphaned technologies. 

Inventors focus on the hammer. Entrepreneurs focus on the nail. These two are rarely the same person. This is why so many of the world’s greatest inventions were not commercialized by their creator, and only found profitable expression long after their lifetime. The question is then: how do you unite the Hammer and Nail space in one team, empowering both sides equally in constant dialogue? When you manage to do that, great things can happen.

I come at this question from the point of view of a trained hammer-maker. People with scientific backgrounds like me get credit in academia for creating more and more powerful technologies, whether or not they address any existing human needs. This, to me, is one of the great challenges for companies working in the AI space. Large language models and image generation models are incredibly powerful technologies — but how do they create value? Are we creating products or just a feature for someone else’s product? The problem with that arrangement goes beyond the possibility of companies going under. When hundreds of billions pour into building hammers without nails — aka “technology without product-market fit” — you soon start to see the contours of a bubble market. 

This technological over-exuberance, leading to inflated valuations and unfulfillable promises, isn’t unique to AI. The same cycle — technological breakthrough, runaway market enthusiasm, a crash, and eventual commercialization and distribution — played out in railroads, the telegraph, the telephone, electric lighting, radio, television, the integrated circuit, and the Internet. Alasdair Nairn’s great book on the topic, Engines that Move Markets, shows the uncanny ways that technological history rhymes. AI’s trajectory so far lines up uncannily with this same rhyme scheme. 

At my company Osmo, we have tried our very best to bridge this essential disconnect between invention and commercialization. This guides how I see my role as CEO. I’m a tech-forward guy, and pretty good at building hammers, but not as attuned to finding nails, and so I’ve had to fully empower people who are. I try to give the utmost agency to my colleagues working on the business end at Osmo to find the best possible applications for our capabilities. Working with our Chief Business Officer Eliot Pence has allowed us to leverage our technology toward solving very specific problems. These are our bread and butter: clean fragrance ingredients that can replace environmentally hazardous ones; new detection technologies that can sniff out fentanyl coming across the border; more effective insecticides that have the potential to save thousands of lives. These practical efforts allow us to pursue even more ambitious projects, like scent teleportation, a powerful hammer with a number of possible nails.

Striking this balance is also a matter of interdisciplinary dialogue. The entrepreneurial minds who work at Osmo need to understand the science to understand what we can reasonably build. This requires a certain amount of storytelling, something scientists are notoriously bad at. Conversely, we try to encourage our engineers and scientists to be able to think entrepreneurially, even if we have to communicate that in the technical language they best understand. 

To the mathematically-minded, transforming a technology into a useful product application can be understood with a little analogy to the Fourier transform, which I call the ‘Hammer-Nail Transform’, an imaginary function that can turn a description of a technology into all of the customer problems it can solve. Applying the (still imaginary) inverse Hammer-Nail Transform takes a customer problem, and shows you all the technologies that can be used to solve it. It’s useful to know whether, at any given moment, you’re thinking in ‘hammer space’, or ‘nail space’. 

This kind of dialogue and translation has helped us find large, single peaks in nail space, and solve them with our unique hammers. Eliot recently summarized our internal approach to finding nails in a series of questions, inspired by the Heilmeier Catechism, which DARPA used to use to pick important problems (like, for example, inventing the Internet). This was for the benefit of everyone working at Osmo, but I think should resonate far beyond that: 

• What are you trying to build? Articulate with no jargon.
• Has it been built before? If yes, go use or copy that.
• Can we uniquely build it, and do other people think it’s impossible? If it’s useful, and we can’t uniquely do this (perhaps that’s time-limited), and people generally believe it’s possible, someone else will make it. Then we can just use it.
• Can this hammer hit many important nails? Can we at least roughly sketch this being useful to the world in a concrete, commercial way?
• If the hammer gets better, can we hit more or more valuable nails? If we go 0->1, and it works, will improving it 1->10 and 10->100 widen a competitive moat, or does its utility level off?
• How much will it cost, and how long will it take?
• What are the midterm and final exams to tell us we’re making progress?

These are the kinds of unifying concerns that have helped us get our Hammer-and Nail-makers working closely together in pursuit of our shared goals. Their combined efforts are helping us build up our company sustainably, creating a solid foundation beyond boom and bust, without sacrificing our ability to dream big. 

‍

‍