Quintin Frerichs

There’s a new kind of biotech company coming out of Silicon Valley. Until now, this has happened mostly by chance, discovered by a few founders independently, without much coordination or explicit top down strategy. Some of the features of these companies are: 

1. They have an expansive, positive mission centered around human biology. Perhaps it’s to live an additional 10 years of youthful life, to unlock new ways to have healthy children, or to dramatically improve our mental health. While sci-fi in its goals and highly technical in its execution, the mission generally has a relatable tie to a global societal challenge, like an aging population, demographic declines or a global “mental health crisis”. 

2. They’re led by an exceptional, often proven founder. In many cases it’s a repeat entrepreneur who’s either already founded or been on the ground floor of a generational company. They bring the credibility, company building skills, and startup instincts needed to succeed in an extremely difficult, long-term, capital-intensive environment. These founders are looking for a transformative technology rather than the typically incremental advances of the industry. 

3. There’s a smooth trajectory from treatment to enhancement: these companies begin by treating a condition that’s responsible for significant human disease and/or suffering, but expand to being available to functionally everyone. The ambition is often to ultimately offer a fundamentally new capability to humans, beyond anything we’ve seen devised by evolution. 

4. They are centered around a novel and promising research kernel, combined with the best the startup playbook has to offer. That means a team composed of the best researchers in a new field with operators and engineers from the strongest startups, who help to set the pace and ambition level of the company bottom-up.
   
   

Even 5 years ago, it’d be hard to point to more than one or two instances of such a frontier bio company. Now there are ~5-10 established and it’s not uncommon to see them raise $100m from world-class investors for heavy R&D and a long-term mission. Working with Fred over the past few years I’ve been fortunate to get involved across the spectrum from investor to founder in several of these companies: New Limit (extending human healthspan), Conception (turning stem cells into human eggs) and Nudge (building whole brain interfaces for everyday life), and each has shown exciting technical progress as well as built strong teams and companies that can last. 

Even with this progress, there hasn’t been a true breakout success in the category. There hasn’t been a company that has converted deep research into a product that helps millions of people, or become a real market success (e.g. $100B+ company). It will likely take years still for that to happen, but the traction is clear and there’s nothing I see as more worth dedicating a long-term mindset to. My goal is to help somewhere between 1-3 of these companies from idea to execution in the coming few years through investing and incubation. 

Ultimately the most important factor in success is pairing a sufficiently compelling mission with a team that can execute over the long run. In effect, the mission motivates the best people to put an absolutely unreasonable amount of effort into the resulting companies. It also shapes not only their success as businesses but their impact on the broader culture. 

I’m dispositionally drawn to a mission that has a focus on both reducing human suffering and enhancing the lives of already healthy people, ultimately including myself, friends and family. There’s something really powerful about being able to draw on each of these motivators, at different parts of the life cycle of the company. It dovetails nicely with what’s best for startups’ success more broadly - first you have to get your company to do *anything at all*, then do something incredibly important to a small group of people, and eventually for everyone. 

In an age largely defined by progress in AI, I think progress in biotech, especially which comes to affect everyone positively (not just those with debilitating conditions) provides an excellent answer to “what all this AI progress is for”. AI leaders have begun to articulate this already, but there’s not nearly enough optimization pressure focused on actually building the projects nor enough active effort to apply AI to the domain in a meaningful way. Drawing more compute and research talent towards bio in the coming years can help resolve one of its core challenges - that of generally long timelines to solving technical problems and slow feedback loops. 

What would a breakout success look like?

The oft-mentioned GLP-1s are a fitting recent example. It’s a public health success, meaningfully treating the societal challenge of obesity, it seems to have broad-ranging health benefits (against addiction, ..) outside of its label and to already ‘healthy’ people. They’ve produced $1T of value for a product that now practically everyone has heard of and changes the way we think about weight loss at a pretty deep, conceptual level. In an industry that often produces failures or only incremental progress, it’s important to recognize that outlier, step function improvements are possible and to shoot for them. 

If we succeed, we should see outcomes rivaling and ultimately exceeding GLP-1s coming from frontier bio startups, as well as truly new scientific breakthroughs typically reserved for academic labs. 

The products they build should feel like a mix of what we love about medicine and what we love about the best consumer products. Maybe it’ll have first saved the life of someone you care about from brain cancer but now it gives you a perfect night of sleep every night. Or be a breakthrough liver treatment until it's making you feel like you're 20 again at 50. 

I’ve now seen what an excellent 0 to 1 trajectory looks like at Nudge and am eager to emulate it across the ecosystem. The confluence of talent required (executive, scientific, engineering, operational) to achieve that is so great that I don’t expect there to be many such opportunities worth pursuing, and I want to reserve attention for the few that meet the bar. The problems at this stage are as such constrained by human capital, and the potential more than justifies the effort. If you’re an exceptional founder, researcher, engineer or operator who wants to get involved in this ecosystem as it hits its stride, please reach out on X or email me at quintin@prometheus.com. 

A list of the best examples to date: Nudge, New Limit, Neuralink, Kernel, Conception, Science Corp, Until Labs, Loyal, Retro

Thanks for feedback on drafts of this post to Fred Ehrsam, Matt Krisiloff, Lucas Harrington, Mackenzie Dion, Blake Byers, Milan Cvitkovic, Laura Deming, Dylan Field, Joanne Peng, and Celine Halioua

Thanks to Blake for the coinage of ‘Human Enhancement Company’ 

I recently co-founded Nudge with Fred Ehrsam, a company with the goal of creating hardware that dramatically improves the daily lived experience of people everywhere. We're using ultrasound to safely and non-invasively measure and modulate brain activity at high resolution — we see huge promise for both those struggling with mental health challenges and eventually people who are generally healthy and want to enhance their everyday life. Ultimately, practically everything that we care about is in some way tied to the quality of our state of mind - and few technologies can interface with it directly, especially in a way that’s beneficial over the long-run. 

While it's hard to predict the full range of possibilities we'll see from our device, we think the highest impact capabilities will be increasing wellbeing, agency over how we choose to spend our time, and the ability to think clearly and learn quickly. While we have a long way to go on delivering these capabilities consistently and at scale, we're starting to see real traction on the science and engineering that would enable such a device and are working in earnest to make it a reality. 

Non-invasive (without a surgical implant) brain stimulation over the past 5 years has already demonstrated the capacity to change people's lives for the better. Magnetic stimulation, namely an optimized protocol of rTMS first developed at Stanford, has shown ~80% efficacy in a randomized-controlled trial at treating intractable forms of depression during only a week of intensive treatment. There are reports of the treatment being transformative for many who participate, a true "before and after" moment. Ultrasound is starting to show the signs of a similar inflection point: we're seeing results of a single, under 30 minute treatment that can enable those addicted to opiates (and who have tried many other forms of standard treatment) to stop using for a month or more. From an article on the first results of the study, patients are reporting long-term effects. While we still need to see how results hold up in a fully randomized, controlled trial, it's an exciting first look at what may be to come:

Part of the power of this technology is that once it demonstrates efficacy for a given treatment, this implies a more generalized set of effects — not only can we target a brain area responsible for the reward learning implicated in addiction, but others which may help treat depression or pain, simply by changing the input parameters and targeting. In fact, we're already seeing treatment of chronic pain in a neighboring region show strong effectiveness in a single 40-minute treatment in a controlled trial. 

There are many ways to interface with the brain, both invasively and non-invasively, but very few are truly safe, scalable, and have compelling use cases. The primary physical ways to interface with the brain are optically (using light to either measure the properties of or affect a neuron's state), magnetically (as mentioned above), electrically (with brain implants like the current brain-computer interfaces used in patients with movement disorders), and acoustically (imaging and stimulating through sound or ultrasound). Optical approaches are limited by the depth light can penetrate into the brain to either implants or surface level brain imaging, potent electrical interfacing requires implantation, and magnetic approaches, while non-invasive, can't be miniaturized and used to target deep brain regions due to physical constraints on the focusing of magnetic fields. Only ultrasound has the overlapping advantages of being incredibly safe on the body (there are decades of research on using low-intensity ultrasound to image fetuses in the womb), can be focused using beamforming to target deep brain regions of millimeters in size, can be used for structural and functional brain imaging, and has hardware that can be miniaturized and scaled without needing an implant. 

While ultrasound's potential as a treatment for mental health diagnoses is just starting to become apparent, we think everyone could eventually benefit from the technology. While over the years I've been blown away by the progress on brain-computer interfaces, faster control of a computer cursor or typing with my thoughts isn't nearly as important to me as my ability to focus and learn effectively, regulate my sleep or stress levels, or explore new states of mind. On the imaging side, I'm excited for brain interfaces that create motifs for communication that are only enabled by understanding circuit-level activity — revolutions in computing happen when the medium creates new gestures for interfacing, rather than just speeding up old ones.  

We have a lot of work to do both scientifically and technically to make the vision of this technology a reality. It's something of an open secret in the field that there are still many hard problems which have plagued researchers for years: incomplete models for the mechanism of action, inaccuracies in acoustic simulation, bulky and/or imprecise transducer hardware, a lack of clear feedback from brain imaging, differences in neuroanatomy or function across individuals, and a lack of definitive parameters to use. Along with the tantalizing recent results in humans for addiction, chronic pain and more, we're starting to gain a foothold in solving many of them in the last 6 months — some solutions coming from the broader research community and some already at Nudge. I ultimately believe that for a compelling product to emerge from this approach we'll need to solve all of these problems at once, and in the same company which has the capacity to scale the product that comes out. If you're an exceptional researcher or engineer and want to work on technically challenging problems and a meaningful mission, please reach out.

Over the past couple of years there have been a number of new neurotechnology hardware platforms developed, from semi-invasive to fully invasive (Neuralink having recently reached their first human implantation), with capabilities for everything from high bandwidth interfacing with the retina to whole brain imaging to targeted neuromodulation. While this is incredibly exciting for the field, there's still a group of people with a potential to impact it who have almost no good routes for contributing: early-career software engineers. 

I've met a number of ambitious young engineers with primarily software skills who want to either start a company in neurotech, build a portfolio project to demonstrate ability and interest in the area, or just make something cool using brain signals. There is basically one option available to them, which is consumer EEG (or in some cases surface EMG, which is a fairly indirect method of collecting brain activity), and it's a really limited one. Even research-grade EEGs, with hundreds of contacts in controlled lab conditions top out at a few cm of spatial resolution (10s of thousands of neurons) and are only reliable for collecting cortical data. There have been some impressive demos using EEG for control but studying the deep brain, getting high bandwidth input and output for high fidelity computer control, discovering new modalities for interfacing with AI, all would either require or greatly benefit from new hardware. The hardware is both years away and likely to come more and more from companies without open API access. 

In the meantime, I think it would be extremely valuable to make simulation environments more available to outside contributors, which was the inspiration behind NDK, the Neurotech Development Kit, which I worked on with AE Studio, Milan Cvitkovic and Sumner Norman. Developing hardware and experimental design within neurotech companies often starts with or heavily uses simulation, and well-documented and open packages allow anyone with software skills to contribute to the space. Furthermore, given the speed of iteration in simulation, I could imagine these packages having an impact on the state of the art similar to what we've seen with robotics and reinforcement learning in the past, where simulation is a key part of making faster progress. In fact, part of the original impetus for NDK came from the success of the OpenAI gym in offering a standard environment to benchmark RL algorithms, which greatly accelerated the field. 

I think some of the most impactful future neurotech devices will be completely noninvasive, and the first use case for NDK is for modeling transcranial ultrasound for neuromodulation. It's open for contributions to anyone who wants to work on neurotech but doesn't have a lab or the hardware, and we hope to see what the world can build with it! 

During 2021 I worked with researchers at the Johns Hopkins CPCR and UNC to develop a new tool for connectome-specific harmonic waves, a technique first introduced by Selen Atasoy at Oxford for measuring whole brain network states in a fairly simple decomposition to around 100 structural/functional 'modes'. We wanted to make the technique usable by a larger number of scientists in the field by open-sourcing the code and analysis pipelines, and making use of more modern techniques from the software development community like containerization. It was fascinating, although not that surprising, to see how much cloud computing, open source neuro data, and improved code made collaboration on the project smoother and the reach broader. The long-term impact, I hope, is that more technical talent in other fields start to see neuro data as a resource for building magical applications in addition to understanding what the brain is doing - the best example I've seen of this recently is "Mind's Eye". 

So-called 'connectome harmonics', introduced in 2016 in this publication of Atasoy's in Nature, are a beautiful way of describing oscillations of activity throughout the brain and have proven useful for characterizing states of neural activity. There are plenty of descriptions of the math underneath connectome harmonics, which borrow from graph theory and show up in other unexpected places, but I often feel the significance of the approach can be lost between visualizations and jargon-filled language of the formal process. In my mind CSHW and related techniques provide a way to simplify the representation of the brain from the activity of 100 billion neurons (plus other emergent phenomena like local field potentials) to 100 dimensions that still give a rich understanding of the brain's state at a given moment in time. By describing such a state as "just" characterized by these 100 or so dimensions, we can further start to think about distance functions between these points (comparing one brainstate to another mathematically), trajectories through this high dimensional space (comparing brainstates over time), and critically, benchmarking such states across different interventions (when I do X to the brain, what trajectory does it follow?). 

I spend much of my time now working on new technologies that will modulate brain activity towards some desired end - be that something relatively mundane like sleep or focus, or more exotic like 'advanced meditation'.  A conventional "functional localization" view of neuroscience might tell us that we should target changing activity in the hypothalamus for modulating sleep or locus coeruleus for modulating arousal but I imagine there are far better metrics for both the endpoint (what is "sleep" or "good sleep", actually?) and more refined strategies for where to target that will fall out of a better understanding of these endpoints. Ultimately, I hope a union of mental models from science, engineering and math all contribute to directly improving people's state of mind.