Author: Sophi Zambakhidze, BRV Fund Manager
Sophi :
Ronjon, reflecting on your transition from being an entrepreneur to becoming a venture capitalist, can you share what motivated this shift? How does your background as an entrepreneur influence your approach as an investor?
Ronjon :
I began my journey in the tech industry after completing a degree in Electrical Engineering, followed by a PhD in Speech Recognition at Cambridge.I pursued further studies at MIT’s Sloan School of Management, focusing on Neural Networks for stock market prediction. My academic path led me to Stanford as a postdoctoral researcher where I delved deeper into artificial intelligence, creating my first AI system back in 1983.
Finding a traditional job proved challenging, and that was the start of my journey as an entrepreneur. My first company was acquired by Motorola, followed by exits to BlackBerry and Apple.
The next step was transitioning from an entrepreneur to an angel investor, I’ve supported around 60 companies, gaining insights and experiences that are invaluable in the venture capital world. This led to a reflective moment facilitated by Stanford’s Distinguished Careers Institute, a program designed for professionals looking to redefine their future after significant careers. The program inspired me to professionalize my investments, culminating in the founding of the R42 venture firm.
At R42, our investment thesis revolves around AI and longevity science—what I like to call “mathematics and medicine.” This focus on deep tech and science reflects a natural progression from being an entrepreneur to an investor. There’s a certain thrill in both “riding the horses” and “betting on the horses.” I’ve found that having hands-on experience in starting or working in a startup is incredibly beneficial for any venture capitalist. It’s one thing to give advice; it’s another to have lived through the startup challenges firsthand—like figuring out how to meet payroll when funds are tight.
Our firm tends to invest in sectors where the technology is tough – deep tech, deep science; and the competition might be less, but the challenge of market development can be greater due to the innovative nature of the products. It’s about recognizing potential in what others might overlook or deem impossible.
I have three personas.
My first persona is a professor at Stanford’s School of Medicine in the Genetics Department, I don’t come from a traditional medical background; instead, I hold a PhD, not an MD. My teaching focuses on the intersections of AI, genetics, and ethics, as well as venture capital and healthcare entrepreneurship, all within the medical school rather than the business school.
Secondly, my role is an investor and managing director of R42, a venture capital firm. We typically invest in startups with promising potential in both the U.S. and the UK, focusing on deploying initial investments of around $250,000.
Lastly, my third persona is an inventor, I continue to be deeply involved in hands-on development. Currently, I’m dedicated to one major project at a time; recently, I’ve been working on SuperBio, which is essentially an AI life science app store offering AI libraries paired with user interfaces. Another ambitious project is AGEMICA, where we’re attempting to develop a vaccine targeting aging—a true moonshot project akin to the kind I’ve tackled throughout my career. For instance, back in 1983, I was working on speech recognition technology, which took about two decades to become mainstream. My approach has always been to set grand goals and methodically engineer toward them, piece by piece.
We’re also addressing sub-problems related to aging, which isn’t classified as a disease by the FDA. This involves creating a platform to develop drug combinations targeting clusters of age-correlated diseases, starting with cancer.
Sophi :
R42 concentrates mostly on biotech and longevity, how did you come up with this particular fund’s thesis? And do you think it’s the next big thing?
Ronjon :
Yes. Well, when I came back to Stanford, really the idea was to look at different problems. What problems are there worthwhile to solve? So I took all the medical school and all the human biology classes. What I discovered was a prevailing mechanistic approach, where traditional observation and trial-and-error methods dominated. However, we’re now witnessing a transformative period where data-driven insights are beginning to revolutionize these fields.
With the advent of the internet about 30 years ago, vast amounts of literature became accessible online, providing a rich resource for data mining. Coupled with databases like the UK Biobank, we now have extensive data repositories that offer valuable insights into genomic states and their interactions with cellular toxicity. Additionally, the emergence of contract research organizations—or what I like to think of as the ‘AWS of biology’—means that startups no longer need to maintain their own wet labs but can outsource experiments globally.
Things are sort of intersecting where they’re becoming useful, where, biology is just beginning to turn into engineering. It’s not quite an engineering subject yet, but you now have tools available to you. And the question is, if you’re starting as a startup, then you can use those tools. Right now, AI is incredibly dynamic—constantly evolving with each passing hour. Looking ahead, I believe longevity will soon become the next focal point in biotechnology. Just like the early days of AI, it’s a field ripe for innovation.
I think the cover of this week’s Economist is the Everything Pill. Consider the recent discussions around drugs like Ozempic, originally intended for Type 2 diabetes but now being repurposed for other uses, despite its side effects. This represents just the beginning of a broader movement where, over the next decade, we expect to see new solutions that mitigate such side effects, further advancing the field of medicine.
Just as AI was a burgeoning topic back in 2010 and has since become a cornerstone technology, I foresee aging science following a similar trajectory, transitioning from a growing interest to a dominant field in the coming years.
Sophi :
With the advent of AI-driven pre-trials accelerating the pace at which new drugs are tested, do you believe the FDA should adjust its regulatory framework to better accommodate these advancements?
Ronjon:
The FDA has been instrumental in safeguarding the U.S. population from harmful drugs. The rigorous process that drugs must undergo to gain FDA approval typically ensures their safety and efficacy for our population. However, with only about 25 drugs approved each year, one might wonder if the regulatory framework is too restrictive, possibly stifling innovation.
On the flip side, there’s an ongoing debate about whether we can fast-track more compassionate use cases. For instance, in situations where patients have limited time, the question arises: can we expedite access to potentially life-saving treatments that haven’t gone through the full FDA approval process? It’s a delicate balance between rapid innovation and ensuring patient safety.
Moreover, current discussions within the regulatory sphere include questioning the necessity and effectiveness of traditional animal testing. The translation of results from animals to humans has been notoriously poor, leading some to argue that we might be missing opportunities where a treatment ineffective in mice could potentially benefit humans.
There are now projects exploring how we might achieve reliable testing results without relying on animal models, which could revolutionize the drug development process. Ultimately, while we’re pushing the boundaries of innovation, particularly with AI, the paramount goal remains the safety of the population. We must ensure that our pursuit of faster drug approvals does not lead to unforeseen and tragic consequences.
Sophi :
Given the current technological landscape, what are some of the barriers in integrating AI with biological research, and how can venture capital play a role in overcoming these barriers?
Ronjon:
Venture capital is essentially risk capital, right? I meet lots of companies that say, ‘Oh, no one’s interested in my company.’ What they really need to do is find a match. Venture capitalists must thoroughly inspect and interact with the founders of biotech companies to understand why they believe their idea will work. It takes 10 years and hundreds of millions of dollars to get a drug to market. Now, with machine learning, can we achieve a higher likelihood of success? Many of the machine learning initiatives have a candidate, but they still haven’t significantly shortened the trial phase yet, which is the expensive and difficult part. What I always ask founders is, never mind the money, they’re going to put ten years of their life into this project. Why this project? What are the signals that this might work, and what might not work? What are the signals that it would work?
That’s what we need to examine. So, I think it will be an interesting discussion. It’s still not a formula yet. Another aspect is the market. What strategies are emerging? One strategy is to go after rare diseases so you can get it fast-tracked. Once it’s approved, maybe it can then be approved more quickly for a larger disease instead of trying to go for the large disease first. What kinds of strategies are now popping up that firms and venture capitalists have to think through? But I’m a big believer in getting things done at lower costs, quicker, faster, cheaper, and safer. If done naively, it’s slower, less safe, and more expensive. And that’s the trick. That’s what sorts the men from the boys, and the women from the girls, in how to actually get something out there as a result.