Reticular AI: The Next Frontier in Drug Discovery

In the world of biotech AI, most companies are playing a sophisticated game of trial and error. But what if you could simply tell an AI model exactly what kind of protein you want to create, just like asking ChatGPT to write a story? This is the ambitious vision that caught our attention at Ritual Capital and led to our investment in Reticular.

Founded by Nithin Parsan and John Yang, Reticular emerges from a partnership that began when the two met as competitors in International Biology and Neuroscience Olympiads. Their paths converged at MIT, where they developed complementary expertise perfectly suited for revolutionizing protein AI. Parsan, an International Biology Olympiad gold medalist, led clinical machine learning projects at the MIT-IBM Watson AI Lab and conducted research at Ginkgo Bioworks. Yang, a mathematics and AI prodigy, made significant contributions to neural network research at MIT’s Media Lab and the Center for Brains, Minds and Machines. Their combined expertise in biological systems, machine learning, and AI optimization creates the perfect foundation for transforming drug discovery.

What makes Reticular’s approach unique is their focus on mechanistic interpretability – a technical way of saying they’re making AI models as controllable as a steering wheel. Instead of the traditional spray-and-pray approach to protein design, Reticular’s technology allows pharmaceutical companies to directly guide these models toward specific biological outcomes.

The company has already demonstrated early success with their technology. In one of their first projects, they successfully steered a Green Fluorescent Protein model to generate more fluorescent sequences by controlling the model’s internal knowledge. This might sound technical, but it’s akin to having a GPS that not only tells you where to go but explains why each turn makes sense.

The potential impact here is massive. The pharmaceutical industry spends billions annually on failed experiments, largely because current AI tools, while powerful, are difficult to control and understand. Reticular’s technology could dramatically reduce this waste while accelerating the development of new therapeutics.

But perhaps what’s most exciting is the timing. We’re at an inflection point in AI development where the same principles that made large language models more controllable are being applied to biological systems. Reticular is at the forefront of this convergence, making protein design as intuitive as having a conversation with an AI assistant.

The road ahead for Reticular is ambitious, but the early results are promising. As they continue to work with pharmaceutical partners and expand their technology’s capabilities, we’re excited to support their mission of making protein design more predictable, efficient, and ultimately, more successful in delivering new therapeutic solutions.

In a field often dominated by hype and incremental improvements, Reticular stands out for its fundamental rethinking of how we interact with AI models in drug discovery. It’s exactly the kind of transformative vision we at Ritual Capital look for in our investments – one that combines technical innovation with practical impact.

Visit Reticular at https://reticular.ai.