Medra announced a $52 million Series A to expand its robotic labs and deploy autonomous discovery systems that pair machine learning with physical experimentation. Backed by Human Capital and Lux Capital among others, the raise targets scaling of engineering teams, lab infrastructure, and validation programs for what the company calls its physical AI scientists.
A New Era: AI and Robotics in Tandem
Medra combines robotics, lab automation, and a predictive layer called Scientific AI to run closed-loop experiments at scale. Robots handle precise liquid handling, assay setup, and data capture. Scientific AI proposes experiments, analyzes outcomes, and updates hypotheses automatically. The result is an integrated cycle where design and execution are tightly linked.
The ‘Physical AI Scientist’ Explained
The ‘Physical AI Scientist’ is a system that couples software and hardware to perform iterative research without constant human intervention. At each cycle the AI generates candidate experiments, the robotic lab executes them, and the measured results feed back into the models. Over repeated cycles the system refines targets, conditions, and molecular choices based on real-world outcomes, not just simulated predictions.
Closing the Discovery Loop: Solving Industry Challenges
Traditional discovery suffers from slow cycles, fragmented data, and weak feedback between prediction and outcome. Partial automation often still leaves critical decisions offline. Medra’s closed-loop approach links prediction and execution so models learn from real experimental results rapidly. That tighter feedback reduces wasted runs, speeds hypothesis testing, and improves the signal for selecting promising candidates.
Accelerating New Therapies for Patients
With $52 million in new capital, Medra can scale robotic capacity and expand its Scientific AI models to explore more biology faster. For researchers and investors this means shorter timelines and more robust preclinical data. For patients the promise is faster identification of viable therapeutic leads and a higher chance that promising biology reaches the clinic sooner.
Medra’s model is not just lab automation; it is a continuous learning loop that aims to make drug discovery faster, more efficient, and more data driven.
