Our mission is curing cancer.

We’re building the biological and clinical data foundation for AI-driven cancer research. Our role is to help researchers turn complex evidence into datasets that support a deeper understanding of disease and the search for better treatments.

We want to help connect what happens in an experiment with what happens in a patient: bringing scientific context to the data used to model biology, investigate treatment response, and guide the next research question.

• Biological and clinical evidence with traceable provenance
• Domain expertise for curation, annotation, and interpretation
• Training and evaluation tasks grounded in cancer research

For teams building models of biology and AI systems for drug discovery, our goal is to make data useful at each step: learning from experimental observations, testing predictions against independent evidence, and identifying where new measurements could help. We’re interested in jointly developing datasets and benchmarks around a partner’s scientific questions.

Engineering depth. Scientific understanding.

Our team brings together experience in model inference, reinforcement learning, and biological research.

We connect scientific expertise with the data needs of AI research, working with contract research organizations (CROs) and biotech companies advancing vaccine development.

Our approach to collaboration starts with a shared research question. Together with data partners and model teams, we aim to define the relevant measurements, organize records into consistent formats, and preserve the protocols, controls, and limitations needed to interpret them.

From biological insight to patient outcomes.

Our oncology focus spans clinical records, imaging, pathology, treatment histories, and longitudinal outcomes. We’re seeking collaborations that connect these patient-level observations with molecular and cellular measurements, helping researchers explore how cancer develops and responds to treatment.

We want to make research data reusable across studies and useful for reproducible model evaluation. That means keeping the evidence behind each example visible, documenting gaps and uncertainty, and defining appropriate uses with the people and institutions contributing the data.

For sensitive data, our approach starts with clear permissions, respect for privacy, and a voice for the people behind the records.

We welcome partnerships with research institutions, AI biology labs, drug-discovery teams, and data partners working toward better cancer treatments. Our aim is to contribute data that can be shared as openly as consent and data rights allow, so more researchers can build on it.