Genesis Molecular AI

Genesis Molecular AI is what it looks like when deep research stops asking for permission and starts compounding advantage in real time. Founded in 2019 by Evan Feinberg, Ph.D., out of Stanford’s Vijay Pande lab, the company didn’t emerge from a trend cycle. It came out of first principles thinking about how molecules behave and how machines might finally understand them. That origin matters because in the startup ecosystem, most companies chase markets. This one is building the map.

At the core is GEMS, the Genesis Exploration of Molecular Space, a system that treats chemistry less like a library and more like a living search problem. Inside it sits Pearl, a 3D diffusion foundation model that doesn’t just predict structure but interprets interaction, pressure, and possibility. This is where the conversation shifts. Not AI as assistant, but AI as generator of viable scientific hypotheses, grounded in physics when real-world data runs thin. Cycles that once dragged now move with intent. Design, simulate, validate, repeat, each pass tightening the signal.

Leadership here is not ornamental. Evan Feinberg operates as CEO with the posture of someone who has seen both the model and the molecule fail and learned from both. Shifeng Pan, Ph.D., as CSO, brings drug discovery discipline where theory meets biology. Alla Ivanova, SVP of Engineering, builds the system to scale without breaking. Sergey Edunov, SVP of Foundation Models, sharpens the engine driving it all, while Aleksandra Faust, Ph.D., as Chief AI Officer, pushes the frontier where model architecture meets application. Paul A. Friedman, M.D., as Chairman, adds the kind of pattern recognition you only earn by getting drugs across the line.

Then the capital hit. A $200M Series B in August 2023 led by Andreessen Horowitz pushed total funding past $280M, not as validation theater but as fuel. In the startup ecosystem, money follows narrative until it follows results. Partnerships with Gilead and Incyte suggest Genesis is already operating in the second phase, embedding into pipelines where outcomes matter and timelines compress under pressure.

What makes this moment real is timing. Drug discovery is expensive, slow, and historically constrained by what can be tested, not what can be imagined. Foundation models, paired with physics, are widening that aperture. Genesis is positioned where compute, biology, and capital intersect, building a loop that feeds itself. Data informs models. Models propose molecules. Molecules generate new data. Each cycle increases leverage.

The company operates across Burlingame, San Diego, and New York, but geography is not the story. Integration is. Engineers sit close to chemists. Models sit close to experiments. Feedback is immediate, and that proximity is where velocity comes from. In a startup ecosystem full of abstraction layers, Genesis is collapsing them.

Genesis Molecular AI is hiring across AI, engineering, and drug discovery. The work is not theoretical. It lands in labs, in pipelines, and eventually in patients. If you are building in AI, biotech, or infrastructure and want proximity to where those lines blur, start at their page. The next wave of drug discovery is not waiting for consensus. It is being computed, tested, and refined right now.